AI News Digest 2026-10-21

特集

DeepSeek、検索とページ取得APIを公開　成功1000回ごとに0.75ドル

[01:58〜]

Muse Glimmer 30B：オンデバイス向けエージェント型モデル
Muse Glimmer 30B: An On-Device Agentic Model

[14:39〜]

React Native 0.88、イベント配送をWeb標準の経路に変更

[26:01〜]

開発者コーナー

GitHub Actions、依存をコミット固定するロックファイルを公開

[38:58〜]

AIツール紹介コーナー

OpenAI、GPT-4oとGPT-4.1を2027年1月14日にAPIから削除

[43:35〜]

速報コーナー

Chrome 155、unloadハンドラを既定で発火させない

[51:17〜]

Apache Camel、HIGH 13件を含む38件のCVEを修正

[52:21〜]

AWS、PyTorchをTrainiumでそのまま動かすTorchNeuronを公開　PrivateUse1の標準バックエンドとしてApache 2.0、10月20日にパブリックベータ

[53:24〜]

Vercel、常駐ワーカーDaemonsのパブリックベータを開始

[54:18〜]

pnpm、発行先と違うレジストリへCIトークンを送らないよう変更

[55:27〜]

Google、Cloud TPU向けネイティブPyTorchバックエンド「TorchTPU」をオープンソース化
Google Open-Sources TorchTPU, a Native PyTorch Backend for Cloud TPUs

[56:31〜]

GitHub、異常時は公開リポジトリのHTTPS取得に認証を要求

[57:38〜]

OpenAI、GPT-6はCOBOLやFortranなどレガシー言語で大幅に向上したと発表
OpenAI says GPT-6 is much better with legacy programming languages such as COBOL and Fortran

[58:36〜]

参考記事一覧を表示

OpenAI、GPT-4oとGPT-4.1を2027年1月14日にAPIから削除 (importance:72)

OpenAI、低資源言語の実装とCOBOL移行で精度を公開 (importance:65)

DeepSeek、検索とページ取得APIを公開　成功1000回ごとに0.75ドル (importance:60)

Hackers hit AI coding tool OpenCode, running up charges on users’ AI accounts (importance:55)

Hackers Use Claude, DeepSeek and Grok in AI-Assisted Attacks on South Korean Banks - GBHackers News (importance:55)

Google Open-Sources TorchTPU, a Native PyTorch Backend for Cloud TPUs (importance:50)

AWS、PyTorchをTrainiumでそのまま動かすTorchNeuronを公開　PrivateUse1の標準バックエンドとしてApache 2.0、10月20日にパブリックベータ (importance:50)

A hacker broke into John Deere and SharkNinja with an army of AI agents (importance:50)

Apache Camel、HIGH 13件を含む38件のCVEを修正 (importance:50)

Chrome 155、unloadハンドラを既定で発火させない (importance:50)

GitHub Actions、依存をコミット固定するロックファイルを公開 (importance:50)

React Native 0.88、イベント配送をWeb標準の経路に変更 (importance:50)

Muse Glimmer 30B: An On-Device Agentic Model (importance:50)

Mercury 3 vs GPT-6 Luna vs Claude Haiku 5.5: Can Diffusion Models Finally Outrun Transformers for Agent Work? (いいね相当スコア: 0) (importance:45)

Advancing coding in low‑resource programming languages (importance:45)

OpenAI says GPT-6 is much better with legacy programming languages such as COBOL and Fortran (importance:45)

Safari、OS 27で識別解決インフラへの通信を遮断 (importance:45)

pnpm、発行先と違うレジストリへCIトークンを送らないよう変更 (importance:45)

Vercel、常駐ワーカーDaemonsのパブリックベータを開始 (importance:45)

Exa 3.0 Search API Tops Perplexity, Brave and Parallel on 8 Benchmarks (importance:45)

TorchTPU - A Hybrid Eager/Compiled Architecture for Native PyTorch Execution on TPU (importance:45)

Training Natural Schemers: RL Can Amplify Oversight-Conditional Behavior (importance:45)

When the harness decides success: state that cannot be re-observed (importance:45)

Does AI-Assisted Development Homogenize Software? An Empirical Study of Code Similarity, Quality, and Evolution (importance:45)

SWE-Bench Pro V2: A Cleaner, Harder-to-Game Coding Benchmark (importance:45)

Continuous Latent Contexts Enable Efficient Online Learning in Transformers (importance:45)

Sequential Behavioral Watermarking for LLM Agents (importance:45)

Mobius Injection: From One-Shot Prompt Injection to Sustained Resource Exhaustion in AI Agent Infrastructure (importance:45)

Agentic Reverse Engineering: From I/O to Code with Discovery Agent (importance:45)

End-to-End Context Compression at Scale (importance:45)

HarnessPro: Recursive Harness Self-Improvement with Pareto Optimization for Performance-Cost Trade-offs (importance:45)

Anthropic’s internal report: 80% of Claude session time is agents waiting on humans (importance:45)

X Bundles Grok, Cursor and Premium Into One Subscription - ALM Corp (importance:45)

Deepseek open-sources the inference engine that powers its fastest and cheapest AI model (importance:40)

NVIDIA Run:ai Is Now Open Source as Part of NVIDIA DSX AI Factory Platform (importance:40)

Voxtral 3 (importance:40)

Tsrs: A Port of TypeScript to Rust (importance:40)

Cutting our inference bill by 68% by being obsessive about optimization (importance:40)

Coding-Agent Benchmarks Should Match Their Users’ Task Flows (importance:40)

“We were looking at it wrong”: The explosive claim that’s upended the AI world (importance:40)

EmbeddingGemma 2 Expands beyond Text to Support Code, Images, Video, and Audio (importance:40)

Mistral AI launches Voxtral 3, a full voice agent stack that listens, thinks, and talks in real time (importance:40)

GitHub、異常時は公開リポジトリのHTTPS取得に認証を要求 (importance:40)

The Window Is in the Reasoning: When LLM Agents Decide to Follow Prompt Injections (importance:40)

When LLM-Generated Refactorings Ripple: An Empirical Study of Repository Adoption Readiness (importance:40)

Arctic-SWE: Revisiting the Long Horizon Agent Training by Modeling-System Co-Design (importance:40)

Break It Down, Pass It On: Cross-Task Skill Transfer in LLM Agents (importance:40)

Improving Environment-Agnostic Agents: The Role of Scaffolding (importance:40)

The Ends Justify the Thoughts: RL-Induced Motivated Reasoning Fools CoT Monitors (importance:40)

Human-AI Productivity Paradoxes: Modeling the Interplay of Skill, Effort, and AI Assistance (importance:40)

Let the Batch Decide, On- or Off-Policy: Effective Sample Size for RL Post-Training (importance:40)

Optimal Post-Training Quantization Scales and Where to Find Them (importance:40)

Service-Induced Congestion in Memory-Constrained LLM Serving (importance:40)

MemPoison: Uncovering Persistent Memory Threats and Structural Blind Spots in LLM Agents (importance:40)

Toollery: Scaling LLM Agents to Thousands of Skills and Tools (importance:40)

Cheap to Hypothesize, Costly to Verify: The Defense Surface of Agentic Vulnerability Discovery (importance:40)

Zero2Repo: Can Coding Agents Build Repositories from Scratch? (importance:40)

COPEX: Benchmarking LLM Robustness to Adversarial Context Across Model Context Protocol (MCP) Layers (importance:40)

Which Model for Which Role? Cost-Efficient Pareto Recommendation for LLM Agentic Workflows (importance:40)

Cognitive Map in Large Language Models (importance:40)

Has GPT-6-Astra Achieved Embodied Intelligence? (importance:40)

US banning Chinese AI DeepSeek from government agency IT systems - Reuters (importance:35)

MiniMax's new M3.1 Flash beats DeepSeek and Opus 5 on a key business benchmark at lower cost — but with a high hallucination rate - VentureBeat (importance:35)

2026 is an exciting year for the Rust compiler (and me) (importance:35)

Publish fake stats, get cited in LLMs (importance:35)

OpenAI acquires Orchestra Bio (importance:35)

ALTK-Evolve: How learned memory tops the τ²-bench leaderboard on customer service tasks without a frontier model (importance:35)

AnyLanguageModel 1.0: One Year of Local Models on Apple Platforms (importance:35)

Nscale and Anthropic are reportedly in talks for yet another $45B compute deal (importance:35)

Microsoft says AI agents are like insider threats as ‘frontier models can’t be trusted’ (importance:35)

Army Cyber Command's ECHOTRIBBLE would turn AI agents into self-spreading cyberweapons (importance:35)

Mistral AI Releases Timestral: A 209M Forecasting Model Family That Outperforms Google’s TimesFM-3 (importance:35)

Cortex: A Fixed-Point Theory of Governed Coding Agents - Technical Report (importance:35)

Who Speaks, Who Authorizes, Who Acts? Understanding Trust Boundaries in Multi-Agent LLMs (importance:35)

Good Judgment, Bad Advice: Decision-Model Controllers Raise Agent Cost Without Improving Success (importance:35)

One Skill Takes Over: Clashing Agent Skills and a One-File Fix (importance:35)

Memory, Know Your PLACE: Governing Lifecycle State and Query-Time Eligibility in Persistent-Memory Agents (importance:35)

Bud ContextCompact 2.0: Context Compaction for Agent Success and Accuracy with Small Language Models and Constrained Context Windows (importance:35)

StopLoss: Teaching Agents to Predict Failure and Stop Early (importance:35)

Design and Cross-Domain Evaluation of an Agent Execution Harness (importance:35)

EvoSkill v2: Fully Agentic Skill Evolution from Benchmark Adaptation to a Stateful Evolver (importance:35)

Honeypots for AI Agent Insider Threats (importance:35)

Same Simulation, Different Evidence: Where Retrieved Tool Output Costs LLM Agents Accuracy (importance:35)

Agentic Interpretation: Harness-Governed Evidence Construction for Software Agents (importance:35)

Red-Teaming Multi-Agent Attacks for Loss-of-Control Risks (importance:35)

Encoder-Free Multimodal LLMs Learn Implicit Encoders (importance:35)

SPIDES: Information-Flow Control for Browser Agents (importance:35)

RATION: Towards Quantitative Risk-Based Access Control for AI Agents (importance:35)

What Matters for NVFP4 Training? A Comparative Study of Low-Precision Pretraining Recipes (importance:35)

Write Once, Check Many: Reusable Ledgers for Post-Hoc Policy Compliance Auditing of Agent Trajectories (importance:35)

When Code Becomes an LLM Call: An Empirical Study of Call-Site Contracts in Agentic Systems (importance:35)

AdamPM: Memory-Efficient LLM Optimization via Moment Pooling (importance:35)

LLM Trojans that Increase Response Lengths and Cost (importance:35)

First-Order LoRA Training with Phone NPUs (importance:35)

What Can't Jev Decide? Probing the Limits of One-Pass Decision Models (importance:35)

Epi: Towards a Persistent, Self-Improving Agent for Autonomous Research (importance:35)

Calibration-Free Early Stopping for Chain-of-Thought Reasoning via E-processes (importance:35)

Safety Offset Learning: When Benign Fine-Tuning Misaligns LLMs (importance:35)

Selecting and Composing Agent Design Patterns: Industrial Case Evidence and Controlled Reference-System Experiments (importance:35)

Beyond Leaderboards: A Method for Application-Specific Capability Profiling of Large Language Models (importance:35)

Anatomy of Off-Policy Policy Gradient: Importance Sampling, KL Regularization, and Baselines (importance:35)

Secure Agentic Coding through Counterexample-Grounded Feedback (importance:35)

When Coding Agents Outpace Review: Evaluating PRD-Driven Development in Mature Enterprise Repositories (importance:35)

Generative to Agentic AI: Survey, Conceptualization, and Challenges (importance:35)

Diagnosing Task Insensitivity in Language Agents (importance:35)

Janus: a Playground for User-Involved Agentic Permission Management (importance:35)

Toward Safe LLM Agents: A Survey of Specification, Verification, and Enforcement (importance:35)

How Good Are Frontier Models at Physics? Expert Re-Grading Reveals Broken Evaluations and Near-Saturation of Leading Benchmarks (importance:35)

PluginRSI: Recursive Improvement of Agent Harnesses with Reusable Plugins (importance:35)

CodeSkill: Latent Skill Abstraction for Long-Horizon Code Agents (importance:35)

RetryBench: Separating Retry Permission from Recovery Outcomes in GPT Tool Use (importance:35)

Empowering Edge Agents: A Systematic Analysis of Edge-Cloud Collaboration in LLM Agent Execution (importance:35)

Compiling Expert Procedures into Verified Programs: When a Small Model Is Enough (importance:35)

Annotrace: A Workbench for Step-Level Annotation of Agent Traces (importance:35)

Bridging Generation and Training: A Systematic Review of Quality Issues in LLMs for Code (importance:35)

Gradient-Free Noise Optimization for Reward Alignment in Generative Models (importance:35)

Probing Persona-Dependent Preferences in Language Models (importance:35)

A Measure-Theoretic Analysis of Reasoning in Transformers: Structural Generalization and Approximation Limits (importance:35)

Your Embedding Model is SMARTer Than You Think (importance:35)

Can Trustless Agents Be Trusted? An Empirical Study of the ERC-8004 Decentralized AI Agent Ecosystem (importance:35)

Automating the Design of Embodied Agent Architectures (importance:35)

Test-Time Augmentation for LLMs: Where to Inject Diversity at Inference Time (importance:35)

Rubric Dropout: A Simple Way to Mitigate Reward Hacking in Rubric-Based RL (importance:35)

Selection, Recombination, or a Fresh Solve? A Candidate-Free Control for Single-Pass Test-Time Aggregation (importance:35)

AURA-Eval: Evaluation Framework for Acting Under Risk Awareness in LLM Agent Trajectories (importance:35)

VReAct: Verification as an Architectural Layer for LLM Agents: A V-Model Design, and a Pilot Study of Its Deterministic Core (importance:35)

OpenCollab: A Multi-Agent Coding Framework with Programmable Collaboration and Controllable Runtime (importance:35)

Principled Under Pressure: Post-Training Decides Whether LLMs Act on Their Own Moral Judgment (importance:35)

PANINI-DR: Controlled Deep Research through Generative Semantic Workspaces (importance:35)

CyberArena: Benchmarking AI Agents in Live Symmetric Attack-and-Defense Combat (importance:35)

SCARCE: Emergent Institutions in Long-Horizon Multi-Agent Coding Systems (importance:35)

PolyGraph: Plan-Buffered CUDA Graphs for Adaptive Speculative Decoding (importance:35)

How Data Scales in Agentic Reinforcement Learning: Laws and Synthesis Strategies (importance:35)

Compaction-Aware Reinforcement Learning For Ascend Triton Kernel Generation (importance:35)

Consistent LLM Call Handling Across Java Frameworks: A Reference Implementation and Benchmark (importance:35)

Parallel Language Models via Variational Decoding (importance:35)

ToolRosetta: Surface-Form Invariance of Tool-Call Guards (importance:35)

Over-Defense in Coding Agents Begins at Silent Forks (importance:35)

Polar Switch: Fast Muon Orthogonalization with Polynomial Circuits (importance:35)

Diet Your LLM: Dimension-wise Global Pruning of LLMs via Merging Task-specific Importance Score (importance:35)

Democratizing Tool Learning with Environments Fully Simulated by a Free 8B Language Model (importance:35)

Train Separately, Merge Together: Modular Post-Training with Mixture-of-Experts (importance:35)

Requests of a Feather Must Flock Together: Batch Size vs. Prefix Homogeneity in LLM Inference (importance:35)

PAAC: Privacy-Aware Agentic Device-Cloud Collaboration (importance:35)

ProCredit: From Outcome Rewards to Progress Credit in Agentic Reinforcement Learning (importance:35)

A Function-Level Vulnerability Score Measures Flag Rate More Than the Model: Protocol Effects on Paired Benchmarks (importance:35)

SP$^2$ec: Adaptive Self-Speculative Decoding for Vision-Language Models (importance:35)

Progress-Aware Distillation for Mitigating Stagnation in Small Language Model Agents (importance:35)

Memory-Induced Tool-Drift in LLM Agents (importance:35)

CVE-2026-106173: FortiSIEM gRPC Path Traversal to RCE (importance:35)

Whoops: Grok Bot Spills Tech CEO’s Bank Data In Public Slack - Techdirt. (importance:35)

「Cloudflare OS」のマネージドサービス提供へ。GitHubとの接続やGmailとの連携強化など (importance:35)

How DeepSeek Is Shaping Huawei’s AI Chips in a Growing Challenge to Nvidia - The Information (importance:30)

Cloudflare's Clairic pulls data from documents faster and cheaper than LLMs (importance:30)

Meta pauses Muse sharing after researchers find AI chats on the web (importance:30)

Spark-X2.6-4B: stronger on-device agents, coding and controllable thinking (いいね相当スコア: 0) (importance:30)

v2.1.304 (importance:30)

MPS in PyTorch 2.15 (importance:30)

GenRecs Tokenization: Transforming Ads Recommendation at Meta with Stable Model Vocabulary (importance:30)

Inside Prometheus: Meta’s Journey to a Gigawatt-Scale Supercluster (importance:30)

Accelerating Mamba-2 with High-Performance Forward Pass Kernels in NVIDIA CUDA Tile (importance:30)

Making the internet better for AI agents might make it worse for everyone else (importance:30)

AI researchers stranded Claude on a virtual desert island. It started to “play.” (importance:30)

Disney found a way to buy Character.AI without buying its lawsuits (importance:30)

ChatGPT now lets you lock your chats (importance:30)

Figma co-founder Dylan Field thinks MCP is not a perfect protocol (importance:30)

Tenstorrent hits $6.2B valuation to scale open-weight AI chips (importance:30)

Elevenlabs releases two new speech-to-speech models designed for real-time AI conversations (importance:30)

Sam Altman reportedly pitched three endgames for AI safety, one involving a single firm melting everyone else's GPUs (importance:30)

CodeQL、macOS 27とXcode 27でトレース解析が使えない (importance:30)

Meet Drex Max: Nace AI’s Multimodal Decision Model for Computer-Use Agents with a Score 67.59 (importance:30)

Doing Less with More: A First-Principles Exploration of the Suitability of Agentic Computing over Alternative Architectural Choices (importance:30)

World of SWECraft: Agent Collaboration at Scale via Simulated Software Worlds (importance:30)

Agents' Last Doubt: Knowing When to Act and When to Step Back in Tool-Using Agents (importance:30)

Probing Ambiguity: Adversarial Intent Resolution for Robust Text-to-SQL (importance:30)

AM-Bench: A Unified Taxonomy and Evaluation Suite for Agentic Misalignment (importance:30)

Hippo: Training-Free Agent Self-Evolution via Hippocampal Replay (importance:30)

HSPR: Recovering Software-Agent Benchmark Scores via Hierarchical Posterior Inference (importance:30)

Selm: Self-Evolving Library Management for Multi-Skill Agent Optimization (importance:30)

Where Natural Language Ends in Agents: Executable Skill Composition with MetaSkill (importance:30)

Teaching Code Agents to Ask Without Leaking Private Code (importance:30)

DAGLoop: Aligning Method Design with Experimental Feedback in Automated AI Research (importance:30)

Editing as Language: Teaching LLM Agents a Grammar for Native Context Editing (importance:30)

Symbolic Constraint Agents: Mining Formal Tool-Call Constraints to Prevent Agent Errors (importance:30)

Whose Credential Is It? How Owner Attribution in Agent MemoryShapes Cross-User Account Misuse (importance:30)

OpenSaC: A Python Library for Search as Code in Deep Research Agents (importance:30)

When Retrieval Partially Conflicts with Skills: Understanding Procedural Hallucination in Language Models (importance:30)

Understanding When Reflective Prompt and Skill Optimization Helps on Locally Served Reasoning Models: A Two-Model Study of GEPA and SkillOpt (importance:30)

CUA-Lite: An Open Platform for Computer Use Agents (importance:30)

VibeTraining-Bench: Data-Trainer-Harness Co-Design for Iterative System Improvement (importance:30)

BOSS: Token-Budgeted Online Sampling and Selection for LLM Reasoning (importance:30)

Revision-Based Learning from Interactive Experience and Feedback (importance:30)

Towards Robust Model Organisms for Eliciting Latent Knowledge (importance:30)

LLM Evolutionary Search Strategies Should be Compared by Tokens (importance:30)

LLM Evolution in a Simple Simulated Market (importance:30)

Unbounded Losses, Unchecked Inputs: Measuring the Safeguards of LLM Options-Trading Agents (importance:30)

Koi: Evidence-Guided, Self-Calibrating Planning for Multi-Cloud, Heterogeneous LLM Inference Fleets (importance:30)

From Stories to Features: Scaling Concept Vectors for AI Alignment & Interpretability (importance:30)

How Much Do Activation Monitors Benefit from Original Model Computations? (importance:30)

Pair-Aware Optimization: Coordinating Exploration and Refinement for Jailbreaking LLM Agents (importance:30)

Robust Direct Preference Optimization under Preference Label-Flip Attacks (importance:30)

PRECOCIAL: Pretraining with Code Curriculums via Iterative Search for Accelerated Learning (importance:30)

UpLoop: Upcycling Large Language Models with Gated Loop (importance:30)

LLM Agents as Suppliers of Physical System Models: Calibration and Acceptance in Modelica (importance:30)

One Head Away from Language Confusion: Single Attention Heads Across Model Families (importance:30)

Policy Is Not Capability: Some Safety Benchmarks Measure Which Definition of "Unsafe" You Trained On (importance:30)

Compile Once, Run Cheap: Frozen, Portable Skills for Embodied Agents, and What Recursive Self-Improvement Adds (importance:30)

Assessing the Empirical Evidence for Loss of Control from Agentic General-Purpose AI (importance:30)

CD-OPD: Contextual Digest for On-Policy Distillation (importance:30)

MemDerive: A Diagnostic Benchmark for Agent Memory over Evolving Data (importance:30)

MOSAIC: Multi-Objective Search for Allocating Inference Cache Across LLM Decoder Layers (importance:30)

DiaLong: Benchmarking Reminiscence in Long-Context Dialogues (importance:30)

ATEST: Adaptive Resource-Aware Test-Time Scaling for LLM Inference (importance:30)

No Matter Who You Are: Addressivity and Sycophancy in Large Language Models (importance:30)

A Calibrated Instrument for Detecting Disposition Drift in Language Models (importance:30)

Computation, Not Memory: Chain-of-Thought and State Tracking in LLM Agents (importance:30)

The KV-Surgeon: Compensation-Aware Offline KV-Cache Compaction (importance:30)

Auditor-Assisted Summary-Channel Verification for Hosted LLM Identity Substitution (importance:30)

NIOCIS: Typed-Decision Next-Hop Routing with Node-Conditioned Option Spaces for Tool-Use Agents (importance:30)

The Development of Sycophancy in Language Models: Tracing Vulnerabilities from Pretraining to Post-Training (importance:30)

Distinguishing Performance From Competence in Evaluations of Humanlike Abstract Reasoning (importance:30)

AI Evaluation Should Require Standardized Item-Level Data Releases (importance:30)

Gypscie: A Cross-Platform AI Artifact Management System (importance:30)

ADIAS: Automated Design of Interactive Agentic Systems (importance:30)

Cascade Failure Attribution for Multi-Agent Systems (importance:30)

When Agent Memory Needs to Compute: Evidence Access and Executable Aggregation (importance:30)

HTN-Orchestrate: A Hierarchical Task Network Framework for Multi-Agent Orchestration (importance:30)

At the Limits of Computability: Steps Towards Autonomous AI Discovery on the Busy Beaver Frontier (importance:30)

SoK: Systematizing LLM Prompt Security: Taxonomies, Datasets, and Unified Evaluation of Attacks and Defenses (importance:30)

Automated structural testing of LLM-based agents: methods, framework, and case studies (importance:30)

Agentic AI -- Physicist Collaboration in Experimental Particle Physics: A Proof-of-Concept Measurement with LEP Open Data (importance:30)

Upgrade the Retriever First: What an LLM Call per Chunk Buys in RAG (importance:30)

Foundations for Investigations of Agentic AI Systems from the Forensic Analysis of OpenClaw (importance:30)

Smart Picks in the Dark: Towards Efficient RLVR for Reasoning via Tracing Metacognitive Pivots (importance:30)

'Your AI Text is not Mine': Redefining and Evaluating AI-generated Text Detection under Realistic Assumptions (importance:30)

Non-Gaussianity Reveals Interpretable Directions in Language Models (importance:30)

The Signs Were Always There: Training-Free Detection and Control in Raw Transformer Dimensions (importance:30)

Evaluating the Robustness of Proof Autoformalization in Lean 4 (importance:30)

From Solvers to Research: Large Language Model-Driven Formal Mathematics at the Research Frontier (importance:30)

SOS-LoRA: Static Orthogonal-Subspace Low-Rank Adaptation with Fixed Multi-Scale Scaling (importance:30)

ISO: An RLVR-Native Optimization Stack (importance:30)

Is Serving-Time Language-Model Inference Necessary for Recommendation? Evidence from Generative and Agentic Settings (importance:30)

CoSec: Benchmarking Agent Security in Communities (importance:30)

RLTLDR: Self-improvement by Internalizing Self-generated Feedback (importance:30)

Learning Self-Orchestration in Decentralized Agent Swarms (importance:30)

UltraData-Code: From Raw Repositories to Algorithmically Dense and Task-Oriented Code Data (importance:30)

Predicting CVSS and CWE from CVE Descriptions with Fine-Tuned Small Language Models (importance:30)

CoEvo-RSI: When Harnesses Evolve, Which Models Should Follow? (importance:30)

Memory as Belief: Bayesian Evolution of Memory for Self-Evolving LLM Agents (importance:30)

Trackable Skill Library Lifecycle Management for Agentic Reinforcement Learning (importance:30)

Nested Shared-Operator Recursion on ARC-AGI-1: An Empirical Study of HRM, TRM and URM (importance:30)

Copying Under Repetition: From Prefix Matching to Position Hashing (importance:30)

Do LMs Know What Variables Do? Interpreting Code Representations via Variable Roles (importance:30)

What RL Brings to Reasoning: A Few Decisions the Base Model Would Not Make (importance:30)

When Do Repetition Loops Lock? Truncation Sampling, Bounded Penalties, and a Blind Spot in DRY (importance:30)

SoK: PIAtlas: Decomposing the Unbounded Prompt Injection Attack Space (importance:30)

From Seeing to Reporting: A Theory of Reliable Visual Set Generation in Vision-Language Models (importance:30)

SPLASH: Contrastive Audit Pressure for Detecting and Localizing Security Hallucinations in Generated Code (importance:30)

Differentiable Mixture-of-Agents Incentivizes Swarm Intelligence of Large Language Models (importance:30)

SignMuon: Communication-Efficient Distributed Muon Optimization (importance:30)

Local Conflict, Targeted Repair: A Perturbation Analysis of Multi-Domain RL (importance:30)

Eliciting Weak-to-Strong Generalization with On-Policy Reverse Distillation (importance:30)

The Complexity Kink: A Prompt-Side Structural Complexity Index for Code-Generation Reliability (importance:30)

ID Balancing: Stable Training of Extremely Sparse MoE via PID-Based Load Control (importance:30)

Invent a Dataset: Measuring Dataset Generation Abilities with Zero Seed Data (importance:30)

Differential Refresh of Materialized Derived State: From a Single-Age Equivalence Limit to an Optimal Per-Partition Allocation, with a Real-System Validation on Materialized Views (importance:30)

EdgeRock: Strong Reasoning at Constant Memory (importance:30)

How Effectively Do Security Skills Guide Coding Agents in Vulnerability Detection in Linux Kernel? (importance:30)

[Paper] REMORY: Learning Residual Memory for Context Compaction (importance:30)

[Paper] DLite: History-free Speculative Decoding for Agentic AI (importance:30)

Musk Swaps Grok Parameters for a 4-Model AI Router [2026] - https://tech-insider.org/ (importance:30)

カスケード型音声対話で宿泊予約の電話応対デモを作る〜投機的実行で待ち時間を減らす〜 (importance:30)

Pocket TTS Small: CPU voice cloning with models as small as 1.9M parameters (importance:25)

FairStress: A deeper evaluation of fairness in LLM decision-making (importance:25)

Why We Still Need an IDE (importance:25)

Dynamics of Sentiment and Attitudes Toward AI Among Developers (importance:25)

Meta tries to scapegoat staff who downloaded torrents to spare execs (importance:25)

Otter.ai raises $320 million from General Catalyst’s growth fund for its meeting tool (importance:25)

Embedded AI evaluators won’t be an easy solution, Hugging Face’s policy chief says (importance:25)

CoreWeave’s debt is becoming the whole AI industry’s problem (importance:25)

Kling AI drops version 4.0 with stereo audio, multilingual lip sync, and a lighter Flash variant (importance:25)

GardenBench: A lightweight, daily evaluation of LLM capabilities (importance:25)

Persistent Memory and Iterative Improvement in an LLM Agent for Chinese-Character Drawing (importance:25)

Cache2LoRA: Understanding and Improving Context Internalization (importance:25)

A Conditional No-Hair Theorem for Transformers: How Language Models Forget Their Prompt (importance:25)

Training-Free Prompt Selection for Efficient Reinforcement Learning With Partial Rollouts (importance:25)

SteerSearch: Interactive Search Reasoning with Mid-Response Steering (importance:25)

Not Every User Says It All: Training Tool-Using Agents Beyond Ideal Simulated Users (importance:25)

The Judge Can Debias Itself: Awareness and Self-Debiasing for Reliable LLM-as-a-Judge (importance:25)

FlashRT-PAI: Fast Kernels Are Not Fast Pipelines for Physical AI Inference (importance:25)

WAQS: Weight Absorption for Quadratic Probing and Affine Steering (importance:25)

Beyond Next-Screen Prediction: The Decision Value of GUI World Models (importance:25)

Train Locally, Learn Together: Distributed Multi-Agent Training via ADMM (importance:25)

Is This Your Agent? Identity Watermarking for Agent Systems in Dynamic Environments (importance:25)

Toward Collective Model Intelligence: Characterizing and Coordinating Models (importance:25)

Bellman Language Models: Solving the Recurring Decisions of LLM Agents (importance:25)

A Cross-Domain Evaluation of Prompting Techniques for Large Language Models (importance:25)

MemPrice: Online Memory Allocation via Adaptive Pricing for Long-Horizon Agents (importance:25)

Toward Agentic Retrieval-Augmented Generation for Enterprise Retail Workflows over Heterogeneous Data Sources (importance:25)

Explore or Confirm? LLM Agents Misallocate Measurements Under Noise (importance:25)

Outcome-Anchored Credit Reallocation for Agentic Knowledge Graph Question Answering (importance:25)

System X: An LLM Decision Model That Replies from Its Own KV Cache (importance:25)

P(Yes): 15$\times$ Improvements via Test-Time Scaling of Decision Models (importance:25)

Auditing Tool Traces in Agentic LLMs: Source Localization and Counterfactual Dependence Testing (importance:25)

Narrow Decision Models: A System-1 Architecture for Calibrated, Typed Decisions (importance:25)

LIRQuant: An Efficient Locally Isolated Rotation Quantization Method for Large Language Models (importance:25)

When Safety Fails Collectively: Robust Alignment Against Harmful Fine-Tuning via Joint Weight Perturbations (importance:25)

ToolOPD: Credit-Aware On-Policy Distillation for Tool-Use Agents (importance:25)

EvalRx: Agentic Diagnosis and Verified Repair of Foundation Models (importance:25)

From Trace to Fix: Residual Task State in Long-Horizon Agents (importance:25)

FrontierRefactor: Benchmarking Multi-Scale Performance-Oriented Codebase Refactoring (importance:25)

MASIF: Implicit Failures for Process Observability in Multi-Agent Systems (importance:25)

Are Self-Evolving Agents Continual Learners? (importance:25)

Claim Laundering: When Agent Decision Models Trust Records Over Reality (importance:25)

From Code to Playable Systems: CEGame-Bench for Long-Horizon Game Creation Agents (importance:25)

The Scheduler Is a Hidden Retriever: How Asynchrony Shapes Evidence Acquisition in Search Agents (importance:25)

Selective Transfer through Sequence Distillation: From Benchmark Knowledge to Backdoor Behaviour (importance:25)

GAIDA: Generative AI for Human-in-the-Loop Document Annotation (importance:25)

Answer Delivery Under a Shared Token Budget: An Audited Study of Stopping and Finalization (importance:25)

Remembering Has a Price: Learning Memory Worth under Resource Constraints (importance:25)

VeriCUA: Learning Proactive Verification for Reliable GUI Agent Rewards (importance:25)

RouterWorld: Scaling Conversational AI with Latency-Aware LLM Routing (importance:25)

Breaking Diffusion: Parallel Reasoning with Draft Refinement (importance:25)

Tandem: Serving Hybrid Linear-Attention GUI Agents on Mobile NPUs (importance:25)

AllPath: Layered Verification for a Deployed, Self-Revising LLM Trading Agent under Human-Gated and Autonomous Regimes (importance:25)

SOMA: Continuous Monitoring and Intervention for System Two Reasoning with System One Models (importance:25)

Agent World Model in the Wild: Computer-Use Agent Environment Synthesis (importance:25)

Multi-LLM Systems Exhibit Robust Semantic Collapse (importance:25)

Anytime Contracts for AI Services (importance:25)

miniTE: Diagnosing and Bridging the Execution Gap in Native NVFP4 Reinforcement Learning (importance:25)

Factorization-Guided Training and Inference for Diffusion-Drafted Speculative Decoding (importance:25)

PeekKV: The First Content Token Already Knows Which Heads Matter (importance:25)

LoopHIL: Looping Hardware into LLM-assisted Agents for Wireless Embedded Systems (importance:25)

Towards Reinforcement Learning for Space Robotics: Fast Training and Reliable Real-World Transfer (importance:25)

VERAS: Verification-Guided Adaptive Search for Reliable Operations Research (importance:25)

Which Long Documents Matter? A Controlled Study of Data Selection for Context Extension (importance:25)

DiffuPlan: Editable Planning State for Diffusion Language Model Agents (importance:25)

AI Rights with an Off-Switch: Rights-Balancing and Due Process Harmonize AI Rights with AI Safety (importance:25)

Same Data, Different Bytes: Serialization-Invariant Integrity Checks for Agent-Published Datasets (importance:25)

ControlJEPA: Principled Trajectory Regularization via Lyapunov Tube Loss (importance:25)

ReTrace: Visual Backtracking for Efficient Multimodal Large Language Models (importance:25)

Auto-MD-bench: Evaluating Autoresearch Agents for Real-World Medical Tasks (importance:25)

NeuroAct: Advancing EEG Analysis and Decoding via Dual-Loop Autonomous Research (importance:25)

BatchParScale: Sparse Batch-Mixing for Sequence Models (importance:25)

EG-GRPO: Evidence-Guided Step-level Credit Assignment for Deep Search Agents (importance:25)

Where and Why Search-Augmented LLMs Fail When Reasoning over Rapidly Evolving Knowledge (importance:25)

Formally Verified Autoresearch for Analyzing the Expressivity of Multi-Head Attention (importance:25)

Weighted State Souping for Efficient Long-Context Retrieval (importance:25)

MapSteal: Map-based Autonomous Knowledge Stealing Attack Against Large Language Models (importance:25)

Consistency Training Suppresses Cue Following Without Teaching Cue Selectivity (importance:25)

Diffusion Language Models Change Speed Across Tasks and Languages (importance:25)

Where RL Breaks: Localizing and Repairing Failures via Targeted Continued Pretraining (importance:25)

Measurement Validity and Representation Dynamics of Confidence Signals in Large Language Models (importance:25)

Adaptive Decoding Methods for Token-Efficient Reinforcement Learning (importance:25)

Policy Search in Code Space for Multi-Agent Pathfinding (importance:25)

AdaExploreBench: Benchmarking LLM Agents for Proactive Exploration and Test-time Adaptation in Evolving Environments (importance:25)

Large-Scale Biomedical RAG: Corpus Scaling, Document Representation, and Efficiency Trade-offs (importance:25)

Fake News, Real Crashes: Information Injection Attacks on LLM Financial Trading Agents (importance:25)

Rethinking LLM Fine-Tuning via Weight Space Reparameterization: Preserving Safety during Downstream Adaptation (importance:25)

Concerns in Benchmarking Jailbreaks: An Empirical Study of Attack Success Rates (importance:25)

TokenGate: Adapting LLMs for Fast and Memory-Efficient Prefill with Learned Token Pruning (importance:25)

RefCraft: Benchmarking Visual-Reference-Conditioned Agentic Generation through Web Development (importance:25)

OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory (importance:25)

Empirical Analysis of Decoding Biases in Masked Diffusion Models (importance:25)

Can LLM Agents Respond to Disasters? Benchmarking Heterogeneous Geospatial Reasoning in Emergency Operations (importance:25)

Beyond Cooperative Simulators: Generating Realistic User Personas for Robust Evaluation of LLM Agents (importance:25)

Herculean: An Agentic Benchmark for Financial Intelligence (importance:25)

Conflict-Aware Multi-Agent Reasoning via Signed Graph Modeling (importance:25)

Distilling LLM Feedback for Lean Theorem Proving (importance:25)

Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization (importance:25)

Defense via Consensus: Mitigating Subliminal Learning through Multi-Teacher Distillation (importance:25)

When the Judge Should Not Decide: Evidence-Locked, Non-Compensatory Selection Bounds LLM-Judge Failure in Reasoning Pipelines (importance:25)

InfraBench: Evaluating Infrastructure Agents Across Layers, Lifecycle, and Risk (importance:25)

When Entropy Is Not Enough: Reclaiming Lost Semantics in LLM Output Length Prediction (importance:25)

Escaping Reasoning Basin Collapse with History-Biased Search (importance:25)

Multimodal Resource-Exhaustion Attacks on Vision-Language Models via Joint Pixel-Prompt Optimization (importance:25)

MemForest: Efficient Agent Memory Management via EventTree Partitioning and Progressive Merging (importance:25)

Taming CoT Obfuscation in VLMs: From Mechanistic Evidence to Activation-Level Enforcement (importance:25)

ActiveMem: Dynamic Latent Memory Trees for Long-Horizon Agents (importance:25)

Finding Side-Effect Loopholes in Verifiable Rewards (importance:25)

How Do Agentic LLMs Decide to Call Tools? A Tool-Call Vector Shaped by Suppression (importance:25)

Beyond Domain-Routed Supervision: Cross-Teacher Guarding for Multi-Teacher On-Policy Distillation (importance:25)

FDE-Bench: Evaluating End-to-End Delivery from Underspecified Real-World Business Requests (importance:25)

Trust Hinders Bidirectional Human-AI Alignment (importance:25)

Input-Conditioned Search for Efficient Test-Time Scaling of Recursive Reasoners (importance:25)

Read the Room Before Your Model Generates (importance:25)

SwarmRSI: Recursive Self-Improvement through Evolving Scientific Agent Swarm (importance:25)

Planning Commitment of LLM under Partial Observability (importance:25)

Do Agents Need a New Kind of Paper? SPRA: Structured Paper Representation for Agents (importance:25)

Realtime-Bench: Can Frontier Models Solve Real-time Interactive Problems? (importance:25)

AutoMind: Adaptive Knowledgeable Agent for Automated Data Science (importance:25)

Open-sci-ref-0.02: open reference baselines for reproducible language model and dataset comparison (importance:25)

Test-Time Compute Games (importance:25)

MOSAIC: More LLMs, Less Cost through Collaborative Reasoning for Compiler Optimizations (importance:25)

SHINE: A Scalable In-Context Hypernetwork for Mapping Context to LoRA in a Single Pass (importance:25)

RAG Strategies for Natural Language-Based SQL Query and REST API Call Generation (importance:25)

VLA-Pro: Cross-Task Procedural Memory Transfer for Vision-Language-Action Models (importance:25)

Balancing Image Compression and Generation with Bootstrapped Tokenization (importance:25)

Is LLM's Moral Decision Compositional? (importance:25)

Circuit Claims Depend on What Is Extracted and How It Is Compared (importance:25)

KAYROS: An Anytime and Exact Open-Source Solver for Duration-Minimization Time-Dependent Vehicle Routing. A Technical Report and a Case Study in Human-AI Engineering (importance:25)

EditPPT: Structured Execution and Evaluation for Reliable Slide Editing (importance:25)

MathAdv: What Theorem Provers Know, Reason, Formalize, and Generalize (importance:25)

ICON decomposition: Multivariate concept auditing reveals shortcuts that neural networks rely on (importance:25)

Good Memory Has ECC: Evaluating the Memory of Vision-Language Models Beyond Accuracy (importance:25)

Leveraging Low-Level Symbolic Competences for Unsupervised Grounding in Hallucination Detection (importance:25)

Chopthin-Consensus Power Sampling: A Diversity-Preserving Approach to LLM Decoding (importance:25)

STAR: Adaptive Spatial-Temporal Normalization for Unified Microservice Incident Management (importance:25)

DashAct: A Progressive Diagnostic Benchmark for GUI Agents in Interactive Dashboard Analysis (importance:25)

SLDBench v2: Active Scaling Law Discovery (importance:25)

Genesis Steering: Better Activation Steering Through Training Data (importance:25)

A Jury, Not an Electorate: Label-Free Reliability Estimation for Ensembles of Language Models (importance:25)

DialFork: Going Beyond Factuality to Evaluate Procedural Hallucination in Conversational Agents (importance:25)

LatLM: Line-Parallel Language Models with Deferred Commitment (importance:25)

Dynamical Kernel Attention: Estimating the Present from a Record of the Past (importance:25)

On Reparameterizing the Score Function (importance:25)

Dataset Inference for Data Provenance and Privacy Auditing in Tabular Foundation Models (importance:25)

An important Bottleneck in LLM Optimizers: Principle-Level Local Optima (importance:25)

Muon Operates Beyond the Edge of Stability (importance:25)

Measuring the Task Value of Precision Upgrades in Closed-Loop Agents (importance:25)

CoTs remain monitorable when being unfaithful requires computation: Replicating and Extending the Necessity Effect Across Eleven Models (importance:25)

TideKV: Delayed KV Eviction via Layer-Wise Chunked Prefill for Memory-Efficient Long-Context LLM Inference (importance:25)

FlowMaker: From Tasks to Cost-Efficient Generative AI Workflow Deployments (importance:25)

Conformal Prediction-Enabled Sequence-level Policy Optimization for Reasoning Language Models (importance:25)

Indistinguishable Agents: Training Agent Privacy as Twin-World Indistinguishability (importance:25)

Maestro: Agentic Formal Proof Development by Coordination (importance:25)

Text-to-Image Generative Modeling Without Language Pretraining (importance:25)

VERDICT: Visual Extraction-Result Decoder with Integrated Confidence and Traceable Evidence (importance:25)

Preserving Reasoning Diversity in Reinforcement Learning with Verifiable Rewards via Adaptive Replay (importance:25)

Can This LLM Annotate My Data? Estimating Zero-Shot LLM Performance (importance:25)

Can Agents Keep Secrets They Were Never Told? Benchmarking Latent Privacy Risks Across Sessions with MemJoin (importance:25)

Write Once, Query Many: Evolving Compact Reusable Memory for Long Videos (importance:25)

Many Repairs, One Direction: Steering Code Models Toward Safer Cryptographic API Use (importance:25)

Persistent Representational Misalignment Induces Vision-Language Model Failures (importance:25)

RARE: Reinforcement Learning with Adaptive Rubric Evolution for Open-Ended Generation (importance:25)

Learning to Reason Efficiently with Interventional Step Credit (importance:25)

Internalizing Chain-of-thought into Looped Transformers (importance:25)

Power Distribution Bridges Sampling, Self-Reward RL, and Self-Distillation (importance:25)

Loop-$\mu$P: Residual Scaling and Hyperparameter Transfer in Looped Transformers (importance:25)

Superhuman AI for Generals.io Using Self-Play Reinforcement Learning (importance:25)

RecurTrace: Adaptive Latent Reasoning with Loop-Time Memory (importance:25)

Linear RNN Scaling Laws: When Longer Sequences Beat More Sequences (importance:25)

PALM: Point-in-Time Adaptation for Financial Language Models (importance:25)

User Model Extraction via Belief Self-Distillation (importance:25)

The Dichotomy Between Pattern Recognition and Step-by-Step Reasoning (importance:25)

Reinforcement Learning via Advantage Guidance (importance:25)

How Well Do Vision-Language Models Understand Movies? (importance:25)

SpargeAttention2: Trainable Sparse Attention via Hybrid Top-k+Top-p Masking and Distillation Fine-Tuning (importance:25)

How far do on-premises open LLMs get on Text-to-SQL? A cross-family size and technique frontier on BIRD (importance:25)

Similar Models, Broken Odds: Where Luce's Axiom Fails in Chatbot Arena (importance:25)

SWE-Residual: Evaluating Starting-State Robustness of Coding Agents (importance:25)

Mastodon 5.0: A closer look (importance:25)

Beginning of the Ende (a community Deno fork) (importance:25)

Matt Pocock's Claude Code habits: 5 takeaways about the human, not the AI agent (importance:25)

Writing down our architecture for the agents helped the cold start more than any prompt trick. What goes in yours? (importance:25)

ornith-ai/Ornith-1.5-122B (Qwen finetune) (importance:25)

Phonon is now open-source, 31M and 40M TTS models designed to run on the edge (importance:25)

[Paper] The COUNT: Counting N-grams Unlocks Scaling on Host Memory and Training-Free Adaptation (importance:25)

[Paper] FlashMask-3: Efficient and Expressive Mask-Aware Distributed Attention (importance:25)

Running 35B MoE models on a Tesla K80 (2014) at ~21 tok/s, in plain C (importance:25)

Exa 3.0 (importance:25)

Mintlify Autopilot (importance:25)

Musk Warns Superintelligence Could "One-Shot" Humanity, Same Day Pushes Grok Onto iPhone Side Button - BigGo Finance (importance:25)

UK regulator weighed urgent court action over Grok deepfakes; bar was too high - MLex (importance:25)

doola MCP Brings US LLC Formation to Grok Build, Closing the Gap Between Building an App and Owning a Company - Newswire (importance:25)

xAI Safety — Former Engineer Exposes Grok Risks and Wrongful Termination - The Korea Daily (importance:25)

MELD: Recursive Self-Improvement via Hierarchical Coordination for Multi-Agent Scientific Discovery (importance:22)

Neuro-Inspired Inhibitory Message Passing for Accurate and Token-Efficient Multi-Agent Reasoning (importance:22)

The One-Pixel of Reasoning: One Truthful Relation Is Enough to Mislead LLM Agents (importance:22)

HybridResearcher: Bridging the Structured-Unstructured Boundary for Deep Research Agents (importance:22)

Correctness-Dependent Asymmetric Target Distributions for Self-Distillation Policy Optimization (importance:22)

Mitigating Reward Hacking in Preference Reward Models with Evolving Penalties (importance:22)

LatComm: Reusable Context Compression for Latent Communication between Large Language Models (importance:22)

WeAct: Dynamic Weight Switching as Agent Actions (importance:22)

Safe, or Simply Incapable? Rethinking Safety Evaluation for Phone-Use Agents (importance:22)

Continual Test-Time Learning with Evolution Strategies (importance:22)

Follow the Mixed Groups: Ridge-Tracking Gaussian Sampling for Zero-Variance Web Rollouts (importance:22)

Mining the Self-Rewriting System: A Vision for Data Mining on AI-Native Infrastructure (importance:22)

An Empirical Study of Parameter-Efficient Fine-Tuning for Mixture-of-Experts LLMs (importance:22)

Efficiency Measured in English: Tokeniser Coverage Bounds Small Language Models on Indic Scripts (importance:22)

The Explainability Illusion: Must ViTs Exhibit Patch-Decoding Interpretability? (importance:22)

TempoMoE: Identifying and Steering Fast-Slow Thinking Experts for Efficient Multimodal Reasoning (importance:22)

Geezuz kernel: Qwen3.8-27B at 190 tok/s (620~ tok/s at 8 requests) on two 3090s (importance:22)

Franchisee reconciles cash in 94% less time with ChatGPT Work (importance:20)

AI game clones are making indie developers afraid to show their work (importance:20)

Meta’s Muse hits 10M U.S. downloads in 40 days, a milestone reached by only 11 other apps (importance:20)

Yelp can now call on your behalf to check wait times at a restaurant (importance:20)

Mamdani says feds have abdicated on AI, urges New Yorkers to push back (importance:20)

Spotify pulled 50 million AI slop and spam tracks from its recommendations (importance:20)

Democrats think cracking down on AI could give them an edge (importance:20)

Presentation: Beyond the Prototype: Scaling Framework Agnostic AI Agent Infrastructure with Ray (importance:20)

A Controlled LLM Workflow for Evidence-Grounded Diagnosis of Risk Score Differences (importance:20)

Self-Improving Video-to-Action Models (importance:20)

Distillation, Self-Training, and the Fragility of Learned Chemical Reasoning (importance:20)

IF-MAP: Influence-Function-Inspired Agent Prioritization in Multi-Agent Systems (importance:20)

Bridging Exploration and Persona: Information-Directed Dueling for Test-Time Agent Personalization (importance:20)

HABIT-Bench: Evaluating Habit Induction from Longitudinal Weak Evidence in Agent Memory (importance:20)

Pattern-Driven Evaluation of LLM Code Optimization: Isolating Compiler-Transcendent Semantic Reasoning (importance:20)

CoworkEnv: Synthesizing Training Environments for Long-Horizon Cowork Agents (importance:20)

Towards Human-AI Complementarity for AI Safety (importance:20)

When Form Changes but Logic Doesn't: Building Logic-invariant LLMs through Logical Structure (importance:20)

Stepwise Reasoning Verification: Don't Generate, Classify! (importance:20)

Learning Which Device to Call: Online Tool Grounding for LLM Agents in IoT (importance:20)

Fine-Grained Units, Flat Decisions: Hierarchical Micro-Expert Pruning for Mixture-of-Experts Models (importance:20)

Not Every Rollout Needs Distillation: Correctness-Filtered On-Policy Self-Distillation (importance:20)

Procedural Memory for Language Agents: A Neurosymbolic Perspective on Explicit Goal Structure (importance:20)

Reliable Before Profitable: QuantHarness for Governing Financial Strategy Development with LLM Agents (importance:20)

StrucChunk: Structure-Aware Retrieval for Legal Document RAG (importance:20)

The Control Layer Behind Rule-Following Agents Trading Across Market Conditions (importance:20)

ScopeMem: Latent Memory That Learns Where to Revise Frozen GUI Agents (importance:20)

Agentic Orchestration of EDC Data Space Retrieval and Natural Language Driven Data Processing (importance:20)

BetaMem: Learning to Remember a Changing World for Embodied Agents (importance:20)

From Outcomes to Components: A Counterfactual Evaluation Framework for LLM-Proposed Repairs (importance:20)

LLM-Assisted Bi-Approximate CHC Solving (importance:20)

Web Agent Agentic Reinforcement Learning Decision Model Under Multi-Cost and Failure Risk (importance:20)

ReTTT: Reflection-Driven Test-Time Training via Experience-Based Parametric Memory (importance:20)

The Weight of Teamwork: Does Communication Structure Help LLM Scientists Search for Scientific Laws? (importance:20)

Who Plans, Who Edits? Verified, Cost-Aware Division of Labor for LLM-Driven Analog Layout Repair (importance:20)

Morpheus-2.0: Multiscale Probabilistic Forecasting with Dual-Axis Attention (importance:20)

Tactix: Towards a Programmable Strategy Interface for Flexible SAT Solving (importance:20)

Long Listening Thoughts: Eliciting Open Auditory Reasoning with Deliberative Perception and Cognitive Refinement (importance:20)

Do 3D Large Language Models Really Use 3D Evidence? (importance:20)

SENTRY: A Streaming Edge Network Trust Engine for Real-Time Audio Deepfake Detection (importance:20)

GazeAct: A Human-Gaze-Annotated Manipulation Dataset for Attention-Regularized Policies (importance:20)

Scene-Adaptive VLA: Efficient Autonomous Driving via Dynamic Layer Routing (importance:20)

GLOSS-MT: Controllable Low-Resource Machine Translation with an Editable Dictionary (importance:20)

CorrFlow: Self-Correcting Flow Matching (importance:20)

SpriteStage: Long Horizon Animation Generation via Progressive Scene Programming and Replay (importance:20)

How Well Can LLMs Verbalize Mathematical Formulas? A Multi-Domain Evaluation (importance:20)

CasePlay: Self-Play Reinforcement Learning from Case Reports for Medical Reasoning (importance:20)

TIRE: Top-Down Formula Instantiation with Reliable Execution for Physics Problem Synthesis (importance:20)

Reflection-Guided Dynamics Rectification for Text-to-Image Generation (importance:20)

PLANE: Plug-and-Play Fine-Grained Visual Experts for Hallucination Mitigation in Large Vision-Language Models (importance:20)

Learning without Overwriting: Structural Knowledge Fusion for Continual Learning (importance:20)

Reranking Answers of Large Language Models with Knowledge Graphs (importance:20)

Vision as Proprioception: Estimating Humanoid Robot Pose from Video (importance:20)

World Model Interpretability: From Decoding Representations to Understanding Latent Worlds (importance:20)

Learning When to Abort in Transformer Planning: A Mechanistic Study (importance:20)

Shared-Private Cross-Modal Discrepancy Learning for Imbalanced Multi-Source Audio-Visual Deepfake Detection (importance:20)

Local Evidence, Global Failure: Diagnosing Speech-Vision Temporal Composition in Audio-Visual Language Models (importance:20)

CHSM8: A Chess Transformer that Follows Natural-Language Style Prompts (importance:20)

Aligning Cultural Behaviors in Arabic Language Models via Preferences (importance:20)

Enhanced Masked Diffusion Unlearning (importance:20)

Causal Humility: Governing the Authority of Fairness Claims Under Causal Uncertainty (importance:20)

Rubrics Beyond Evaluation: Exploiting Rubric Structure in Reinforcement Learning (importance:20)

Accelerated Drift of Scientific Prose in the LLM Era: Evidence from an Interrupted Time-Series Analysis (importance:20)

Nash-MOPD: Optimizing Task Trade-offs in Multi-Teacher On-Policy Distillation (importance:20)

Reasoning-Aware Multicalibrated Reliability Predictors for Language Models (importance:20)

Kronecker Delta Rule for Long Context Linear Attention (importance:20)

Markovian Linguistic-Temporal Bridge: Unlocking the Potential of LLMs for Time Series Forecasting (importance:20)

From Language Models to Control Policies: Search, Selection, and Verification (importance:20)

Traceable Context-Enriched Knowledge Graph Generation with Induced Schemas (importance:20)

Schema-Aware Planning and Hybrid Knowledge Toolset for Reliable Knowledge Graph Triple Verification (importance:20)

Virtual Speech Therapist: A Clinician-in-the-Loop AI Agent for Speech Therapy Planning (importance:20)

MOIRA: Policy Memory and Belief Revision for Language-Model Agents in Non-Stationary Pair Trading (importance:20)

Nexus : An Agentic Framework for Context-Aware Time Series Forecasting (importance:20)

MUSE: Benchmarking Assembly Structure and Interface Geometry in Multi-Component Text-to-CAD Generation (importance:20)

Evidence Acquisition and Stopping via Lattice Distillation (importance:20)

Scientific discovery as meta-optimization: a combinatorial optimization case study (importance:20)

Arena-T2I Hard: Benchmarking and Improving Faithfulness with Dependency-Aware Checklist (importance:20)

Harnessing Textual Refusal Directions for Multimodal Safety (importance:20)

Can We Trust Item Response Theory for AI Evaluation? (importance:20)

WIDE: Boosting Adaptive LLM Inference via Token-level Dynamic Width Pruning (importance:20)

MobilePA-Bench: Benchmarking Mobile Planner Agents on Complex Real-World Tasks (importance:20)

Omni-Interactive Universal Embedder (importance:20)

VACS: Value-Aligned Compositional Shielding for Multi-Agent Reasoning (importance:20)

LSTMem: Hierarchical Long Short-Term Online Memory for Large Language Models (importance:20)

Stashbird: Episode-Grounded Lifecycle Reconciliation for Conversational Agent Memory (importance:20)

Synthesis Through Simulation: Generating Coherent Enterprise Data via Scalable Agent-System Interaction (importance:20)

Passing Tests, Missing Semantics: Evidence of Interpretive Lock-In in Plan-Based DataFrame Code Generation (importance:20)

Attention Sinks in VLMs Are Borrowed, Not Seen (importance:20)

HyWe: Unifying Search and Operation for Real-World Hybrid Web Tasks (importance:20)

Recursive Environment Scaling for General Long-Horizon Agents (importance:20)

Jeva: System One Models from First Principles (importance:20)

Competition and Diversity in Generative AI (importance:20)

Real Faults in Deep Learning Fault Benchmarks: How Real Are They? (importance:20)

Toward a Physics of Deep Learning and Brains (importance:20)

IRPM: Intergroup Relative Preference Modeling for Pointwise Generative Reward Models (importance:20)

KernelClaw: A Multi-Agent Framework for GPU Kernel Optimization with Dual-Memory Mechanisms (importance:20)

Brittlebench: Quantifying LLM robustness via prompt sensitivity (importance:20)

Attention Sinks Induce Gradient Sinks: Massive Activations as Gradient Regulators in Transformers (importance:20)

Efficient Training on Multiple Consumer GPUs with RoundPipe (importance:20)

Coral: Cost-Efficient Multi-LLM Serving over Heterogeneous Cloud GPUs (importance:20)

DAMEL: Dual-Axis Multi-Expert Learning for Class-Imbalanced Learning (importance:20)

CANVAS: Captioning Art with Narrative Visual-Audio AI Systems (importance:20)

EDITH: Teaching Robot to Follow Nonverbal Human Intent from Egocentric Vision and Gaze (importance:20)

Arithmetic and its remedies select the evidence on learned geometry (importance:20)

Calling the Bluff: Detecting Ever-Shifting Harmful Chat Dialogue via Ordered Reasoning Chain Regularization (importance:20)

ConfAL-WM: Confidence-Guided Active Learning for Action-Conditioned World Models (importance:20)

GraFT: A Training-Free Framework for Spatial Reasoning in Multimodal Large Language Models via 3D Scene Graphs (importance:20)

From Generative to Discriminative: Lightweight Adaptation of SpeechLLMs for Multimodal Emotion Recognition (importance:20)

The Ups and Downs of Backprop Weights (importance:20)

Post-Training Frontier Text-to-Image Models by Composing Preference and Rubric Rewards (importance:20)

Co-adaptive Rule Routing and Constraining for Structured Knowledge Retrieval (importance:20)

Do Knowledge and Reasoning Oriented Benchmarks Induce Stable Representation Signatures in LLMs? (importance:20)

Alliance of Stochastic Routing and Reinforcement Fusion for Multi-Model Reasoning (importance:20)

Inside the Loop: A Mechanistic Study of Weight-Tied Transformers on Depth-Bound Algorithmic Tasks (importance:20)

ProbePO: Sparse Probes, Redistributed Credit for Critic-Free LLM Reinforcement Learning (importance:20)

EdgeKernelBench: Unified LLM-Driven Kernel Generation for Edge Backends (importance:20)

A Statistical Viewpoint on Adam, Shampoo and Muon (importance:20)

MLFont: Machine Learning-Driven Font Subsetting and Range-Compact Unicode-Range Slicing for Low-Latency Web Typography (importance:20)

Certified Sparse Attention via Perturbed Retrieval (importance:20)

Conjugate Compilation: Self-Improving Agentic Co-Compilation for Speculative Decoding (importance:20)

Beyond the Teacher: Understanding and Preserving Distillation Overshoot in Agentic On-Policy Distillation (importance:20)

PACT: NVFP4 Data-Flow LoRA Fine-Tuning on Consumer GPUs (importance:20)

Improving the Expressive Power of Hybrid Sequence Models via Multi-Hop Retrieval (importance:20)

Where Groups Split: A Probe Prior for Nonzero Within-Group Variance in RL Post-Training (importance:20)

Learn the Interface, Not the Implementation: Output Representation Dominates Training Regime in LLM-Based RTL Generation (importance:20)

Bombadil: A Foundation Model for Plant Electrophysiology (importance:20)

Degree Bounds for Separating Polynomial and Neural Invariants for Point Clouds (importance:20)

PACT: Plan-Augmented Coordination via Trace-Graphs for Dynamic Multi-Agent Combinatorial Optimization (importance:20)

On Data Scaling in Tabular ML: More Data, Better Generalization? (importance:20)

FrameScout: Scouting Query-Relevant Frames for Long Video Understanding (importance:20)

When Does Evidence-Graph Structure Predict Answer Correctness? Label-Free Abstention for Structured Retrieval (importance:20)

Location, Location, Generation: Where You Ask Matters for Generative Engine Optimization (importance:20)

The Missing Compliance Class: LLM Graders That Collapse Jailbreak Success, and a Free Check (importance:20)

An Information-Geometric Framework for OCS-Reconfigurable Interconnect Topology in Large-Scale AI Clusters (importance:20)

LeakBench: Auditing Reward Hacking through Prediction-Time Leakage in Clinical Modeling Agents (importance:20)

Co-Evolving Policy and Reward for Code Generation: A Mechanistic Study of Single-Adapter Drift (importance:20)

See it to Place it: Evolving Macro Placements with Vision-Language Models (importance:20)

Convex Compositional Reasoning Models (importance:20)

A Decision-Theoretic View of Test-Time Training: When, How Far, and Which Directions to Adapt (importance:20)

When Should an In-Context Learner Expand Its Hypothesis Space? (importance:20)

Double-Take: Improving Capacity, Binding, and Selectivity In Long-Context Within Bidirectional Memory (importance:20)

Backdoor Attacks on Discrete Graph Diffusion Models (importance:20)

Language steering in latent space to mitigate unintended code-switching (importance:20)

RF-GPT: Teaching AI to See the Wireless World (importance:20)

Large Language Models over Networks: Collaborative Intelligence under Resource Constraints (importance:20)

What Makes a Latent Thought Valid? Defining and Auditing Thought Representations in LLMs (importance:20)

Flipping Minds by Flipping Bits: Decision-Level Cognitive Hijacking of Large Language Models (importance:20)

Predictable Failure in Multi-Hop Retrieval: Score-Distributional Confidence Scoring and Abstention (importance:20)

From Candidate Admission to Task-Level Return: Auditing LLM-Generated Code under Execution Contracts (importance:20)

Vibenix: Agentic Software Packaging and Maintenance with Nix (importance:20)

Declarative WASM GC Types (importance:20)

Opus 5.5 medium spun up 12 Opus 5.5 medium subagents to read 87 transcripts and burned my entire 5-hour limit in 5 minutes lmao (importance:20)

Who did it better? GPT 6 Astra vs Opus 5.5 (importance:20)

Are your dev teams noticing a massive wave of "AI code debt" and subtle race conditions lately? (importance:20)

Tried Cline for a week after living in Claude Code and Codex, some notes (importance:20)

Qwen3.8-Flash-Next vs Opus 5.5: Six-way benchmark on a real Rust feature with Claude Code and Pi (importance:20)

My obituary for programming (importance:20)

Praex (importance:20)

Intel® GPU D-series (importance:20)

NYC Council grilled SpaceX over a sexualized Grok image of its own speaker - Startup Fortune (importance:20)

Not Just a Selfie: The New BIPA Theory Aimed at Generative AI (via Passle) - Frankfurt Kurnit Klein & Selz PC (importance:20)

Learning to Compose: Recursive Self-Improvement of Shared Executable Stacks for Zero-Shot Long-Horizon Robot Manipulation (importance:18)

When Interpretations Persist: Downstream Influence in Gaze-Aware Assistants (importance:18)

Steering Language Models Before They Speak: Logit-Level Interventions (importance:18)

Visual Text Compression as Measure Transport (importance:18)

Is it RL or Data? A Controlled Comparison of GRPO and Rejection-Sampling Fine-Tuning (importance:18)

From Software History to Executable Regression Tasks: A System for Evaluating Software Agents (importance:18)

From Finding to Fixing: How Useful is Fault Localization for Agentic Program Repair? (importance:18)

Federal Judge Allows NAACP Lawsuit Against Elon Musk’s xAI to Move Forward - The Michigan Chronicle (importance:15)

How 3 Java Champions are using GitHub Copilot for increased productivity (importance:15)

The new economics of enterprise AI (importance:15)

A visit to Samsung’s massive new semiconductor hub (importance:15)

Accused OpenAI spy had folder called “Saudi Gov Docs,” FBI says (importance:15)

Eric Schmidt reportedly owns about half as much of Anthropic as CEO Dario Amodei (importance:15)

Eka is worth $1.45B thanks to a physical AI model that has experts raving (importance:15)

Tolan snags $30M for its AI companion that helps users navigate life (importance:15)

Replika’s founder built the AI assistant she pitched at Disrupt 12 years ago (importance:15)

People are using AI to make their smart TVs dumb again (importance:15)

JB Pritzker thinks AI could help us live to 100 — or threaten us like nukes (importance:15)

What Social Media Teaches Us About Accumulative AI Risks (importance:15)

Unified Multiparadigm Fusion Intelligence for Rare Disease Diagnosis (importance:15)

LogicVal:A Logic-Formalized Benchmark for Constraint Satisfaction in Text-to-Image Generation (importance:15)

Joint, Decoupled, or Agent: When Each Multimodal Alignment Paradigm Wins (importance:15)

From SEC Filings to Financial Signals: Evaluating LLMs on Stock Market Reactions and Analyst Forecasts (importance:15)

Screening LLM Judges Without Ground Truth: A Preliminary Evaluation of an AI Dental Hygiene Coach (importance:15)

DKnown-Jev: Fine-Tuning and Evaluation of a JEV-Style Model for Tasks Involving Normative Knowledge (importance:15)

Pluralistic Alignment in Language Models: Objectives, Human References, and Preference Representations (importance:15)

ConCI: A Controlled Counterfactual Benchmark for Contextual Integrity in LLMs (importance:15)

Axiomatic World Modeling for Physics Reasoning (importance:15)

Comparing Sparsified MLPs and Transcoders: Many Transcoder Features Are Neurons in Disguise (importance:15)

Verifying Human-Written Reasoning References (importance:15)

Are We Evaluating VLM Token Pruning Right? Looking Beyond Accuracy and Token Count (importance:15)

Medical Fine-Tuning Reshapes, Not Reduces, Patient-Facing AI Risk: A Rubric-Based, LLM-Judge Evaluation Framework (importance:15)

Means-Ends Analysis Helps LLMs Plan in Continuous-State Domains (importance:15)

Revealing what benchmark scores hide: Jointly measuring the breadth and reliability of LLM capabilities using accessibility profiles (importance:15)

CorrectRTL: Boost RTL Generation Quality Through Automated Self-Correction with Instruction Decomposition (importance:15)

VLA-Rolling: Fewer Denoising Steps, Better Control via Adaptive Action Reuse (importance:15)

SteadyThought: Mitigating LLM Under-Thinking via Thought-Level Preference Optimization (importance:15)

Spectral Adaptive Graph Transformer for Efficient and Scalable Traffic Forecasting (importance:15)

Dialogue Generation for Conversational Recommendation Using User Search Interactions and Knowledge Graphs (importance:15)

BrandEval: A Two-Track Multi-Agent Benchmark for Risk-Sensitive LLM Crisis Communication (importance:15)

Bridging Concepts: Multi-Agent Reasoning over Scientific Concept Graphs for High-Impact Research Prediction (importance:15)

Aligning Language Models with Indigenous Community Cosmovision through Constitutional AI (importance:15)

On Reasoning Correctness Directions (importance:15)

A failure of Imagination: VLMs fail to use visual representations for text, but self-distillation can help (importance:15)

From Proposal to Execution: Modeling Interaction Semantics in Shared Autonomy (importance:15)

Formal Raising: Embedding Verifiable Ethical Properties into the Development of AI (importance:15)

When Empathy Misses the Goal: A Benchmark for Goal Displacement in LLM Advice (importance:15)

Think, Compress, Answer: Reinforcement Learning for Multimodal Grounding via Reasoning Compression (importance:15)

Spherical Fixed-Point Reasoners: Adaptive Loop Halting on the Hypersphere (importance:15)

Neither Over- nor Under-Constrained: Evaluating LLM Personalization under a Constraint Qualification Audit (importance:15)

CVaR-RMD: CVaR Selection Pressure for Robust Metaheuristic Discovery (importance:15)

From Black Box to Controller: Inference-Time Steering of LLM via Feature Amplification (importance:15)

TimelineAgent: An LLM-based Framework for Query-Centric Event Timeline Extraction Across Multiple Documents (importance:15)

Standing on the Shoulders of Giants: Enhancing LLM Agents for Automated Fact-Checking with Human Expert Skills (importance:15)

A Cue's Hidden Price: Auditing Seller Influence on AI Shoppers (importance:15)

When Should AI Chime In? Benchmarking Timely Contribution of Proactive Agents in Multi-Party Chat (importance:15)

Productive State-Space Exploration for Multi-turn RL in LLM Agents (importance:15)

ABACUS: Program-Guided Multistep Reasoning over Frozen Judgment Models (importance:15)

GLOBE: Geometry-Led Online Behavioral Early Exit via Unit-Sphere Trajectories (importance:15)

ClueFlow: Learning Long-Horizon Search through Attributed Research-State Transitions (importance:15)

AuDQ: Auditable Data Quality through Agentic Rule Discovery (importance:15)

Decision4Rec: A Jev-Inspired Framework of Decision Models on Knowledge and Active Graphs for Multi-Agent Recommender Systems (importance:15)

LLMs Prompted for Legal Context Refuse More: Overrefusal from Small On-Premises LLMs in Criminal Legal Context (importance:15)

A Generic On-Board AI Processing Chain for Earth Observation: From Application Development to Embedded Deployment and Mission-Level Orchestration (importance:15)

Autonomous intelligence for mining productivity: a multi-agent AI approach to operational value creation (importance:15)

Toward Structured Financial Foundation Models (importance:15)

Beyond Behavioral Accuracy: Evaluating Individual-Level Fidelity in LLM Investor Simulation (importance:15)

WhatIf-Bench: Benchmarking Intervention-Aware Time Series Forecasting (importance:15)

StebBench: Diagnosing Real-World Procedural Understanding in AI Agents (importance:15)

ProteinGeoBench: Frontier Structural-Biology Agents Are Fluent at Lookup but Lost in Geometry (importance:15)

AnyGoal: On World Model Generalization Beyond Expert Rollouts (importance:15)

Train An Imaginative Agent (importance:15)

Latent Multimodal Memory Reasoning (importance:15)

EMMA: Event-Driven Multimodal Memory System with Multi-Agent Retrieval (importance:15)

EgoVine: Post-Training with Branching Experience Compiled from Real-World Videos (importance:15)

When the Editing Intent Is Split: A Cross-Modal Jailbreak Attack for Large Image Editing Models (importance:15)

Cross-Modal Forgetting and Transfer in Unified Multimodal Post-Training (importance:15)

SpinDoctor: Finding Innocuous Prompts for Biased Image Generation (importance:15)

Organizing LLM Inference for Aggregate Predictions from Behavioral Data (importance:15)

Beyond Prediction: A Component-Level Study of Concept Recovery in In-Context Learning (importance:15)

How Much Can We Prune? Exploring the Performance Limits of Visual Token Pruning (importance:15)

RL Teaches VLMs Where to Attend but Not How to Perceive (importance:15)

OR-License: Evaluating LLM Agents as Real-World Operations Research Practitioners (importance:15)

Prefix-Tuning for Arbitrary Output Sequences on Pretrained Transformers (importance:15)

From Spectrograms to Shallow Speech SSL: Audio-Frontend Design for Generator-Disjoint and Low-Latency Audio--Visual Deepfake Detection (importance:15)

lepidoptera: A Domain-Agnostic Engine for Causal Simulation with Mixed Variable Types, Vine Copula Dependence, and Individual-Level Counterfactual Propagation (importance:15)

CoDA: A Human-LM Collaborative Codebook Development Assistant for Domain-Specific Text Annotation (importance:15)

Analyzing Repetition and Interference Effects in Memory-Constrained User Simulators (importance:15)

Learning Administrative Boundary-Aware Location Encoders via Reverse Geocoding (importance:15)

Indirect Influences of Artificial Intelligence Advice on Complex Sepsis Treatment Decisions (importance:15)

ZeroDSS: Semantic Segmentation from Class Names Alone (importance:15)

When Opposite Premises Translate Identically: A Source-Conditioned Diagnostic for Multilingual Commonsense Items (importance:15)

HumanSkill-4D: Training-Free Language Querying of Animatable Gaussian Avatars in Space and Time (importance:15)

Tree Plurality Veto: Voting in Few Rounds with Efficient Communication and Low Metric Distortion (importance:15)

Contrastive Hypergraph Source-free Domain Adaptive Object Detection in Adverse Weather Conditions--Extended Version (importance:15)

CARVS: A Congestion-Aware Algorithm for Online Multi-Agent Pickup-and-Delivery in Human-Collaborative Warehouses (importance:15)

ASR Personalization in Pathological Voice: Accuracy, Efficiency, and Duration Effects (importance:15)

Contrastive Adversarial Training for Robust Graph Neural Networks under Label Poisoning (importance:15)

Tensor Network Datamodels for Training Data Attribution (importance:15)

Evidence Field: Adaptive Covering Programs for Medical Imaging (importance:15)

Cooperation Without Shared Rewards via Reciprocal Q-Augmentation (importance:15)

RAFT: Redundancy-Aware Input Filtering for Temporal Graph Neural Network Training (importance:15)

Leveraging Mental Models for Human-Centered Explanations: A Systematic Design Approach Illustrated by Explainable Fake Review Identification (importance:15)

Stable Signer V3: 4D Deformable Primitive Splatting Is All You Need for Sign Language Production (importance:15)

TrustAgent-RFF: Agentic Diagnostic Planning for Reliable RF Fingerprinting under Domain Shift (importance:15)

Label-Space Mismatch Distorts Cross-Cohort Evaluation in Breast Cancer Molecular Subtyping (importance:15)

UtoMe: Observation-Uncertainty-Guided Token Merging for Weather Foundation Models (importance:15)

IECLA: Interpretable Emotion Contribution Learning for Affective Image Editing (importance:15)

Synergy-Aware Residual Flow Matching for Single-Cell Perturbation Prediction (importance:15)

HCFM: Helmholtz Component Flow Matching for Time Series Anomaly Detection (importance:15)

Pre-Query Retention with Bounded Bit and Word Retrieval (importance:15)

Self-organizing reinforcement learning: a swarm approaches the explore-exploit optimum (importance:15)

ClinSDT: LLM-Encoded Clinical Semantic Guidance for Sepsis Treatment via Decision Transformer (importance:15)

M-JEPA: Predictive Self-Supervised Learning for Molecular Graphs with Scaffold-Shift Evaluation on Tox21 (importance:15)

World Models, Trees and Foundation Models for Retail Forecasting: Evidence from Four Public Data Sets (importance:15)

ViCoR: Estimating Visual Necessity via Counterfactual Residuals for Multimodal Medical Data Selection (importance:15)

AdaFedXplain: Adaptive Federated Learning with Explainable AI for Ransomware Detection (importance:15)

Dynamics-Aligned Representation Learning under Induced Partial Observability (importance:15)

Modaliff: A Unified Token Diffusion Model for Any-to-Any Generation (importance:15)

A "feature ODE" describing the learning behavior of shallow MLPs on simple functions (importance:15)

Cross-Pollination Retrieval: Finding Transferable Ideas Across Fields in Scientific Literature (importance:15)

Cognitively Decoupled Meta-Induction for Out-of-Distribution Generalization (importance:15)

Remember Less, Adapt Longer: Bounded-State In-Context Reinforcement Learning with Recurrent Summary Memory (importance:15)

UtilityBench: Measuring Language Model Utility When Errors Have Different Costs (importance:15)

Keep the Lineages Alive: Diversity-Preserving SMC for Diffusion LLM Decoding (importance:15)

Behavioral Probes for Information Flow in LLM Swarms (importance:15)

Don't Rewrite the SID Map: Topology Preservation for Generative Retrieval in Dynamic E-Commerce Search (importance:15)

VEILED-SH: Vulnerability Expression through Indirect Language Evaluation Dataset (Self-Harm) (importance:15)

Intra-Rubric Reward Divergence: Mitigating Rubric Hacking through Latent Dimension Recovery and Balancing (importance:15)

Where is Physical RSI capability from? An Observational Scaling Law of LLM-as-policy (importance:15)

Reinforcement Learning with Visually Attributable On-Policy Self-Distillation (importance:15)

SELECT: Semantic Erasure Boundary Construction with Sibling Exclusive Counterfactuals (importance:15)

Reflection-Driven Self-Improving Language Model Agents for Adaptive RAN Slicing (importance:15)

RADD: Retrieval-Anchored Discrete Diffusion for Multi-Modal Knowledge Graph Completion (importance:15)

Benchmarking the Limits of In-Context Reinforcement Learning for Ad-Hoc Teamwork (importance:15)

VLA-Trace: Diagnosing Vision-Language-Action Models through Representation and Behavior Tracing (importance:15)

Planning Takes More Than Token Prediction: Causal Plan for Benchmarking and Building Physically Grounded Embodied Reasoners (importance:15)

Frame-Conditioned Moral Computation in LLaMA 3.1-8B-Instruct: An Exploratory Mechanistic Interpretability Audit of Ethical Reasoning (importance:15)

MedEvoEval: Evaluating Continual Evolution of Doctor Agents through Simulated Clinical Episodes (importance:15)

NeurOWL: An LLM-Based Neural-symbolic Framework for Incomplete OWL Ontology Reasoning (importance:15)

CausalNav: Reliability-Certified Causal World Models for Control under Physical-Parameter Shift (importance:15)

VERDI:Verified Contradictions for Continual World-Model Optimization (importance:15)

6.5% of the Neuro-Symbolic Literature Can Be Reproduced from Its Published Artifacts, a Six-Stage Audit Framework and First Instantiation (importance:15)

LLM-Driven Autonomous Vehicles Can Inherit Human Driver Biases in Pedestrian Yielding: Results and Implications From a New Benchmark (importance:15)

LayerRoute: Action-Conditioned Mixture-of-Layers Routing for Vision-Language-Action Policies (importance:15)

Multi-Agent Agentic Graph Learning via Structural Signatures (importance:15)

PotARCin: Multi-Dimensional Evaluation of Skill Acquisition in Abstract Reasoning Tasks (importance:15)

Hyper Algorithm Design Agent: Evolving Learnable Optimizer from Zero (importance:15)

FineSID: Scalable and Efficient Semantic Identifier Learning for Generative Recommendation (importance:15)

Agentic AI with Structured CoT for Enhancing AI's Spatial Intelligence: Visualization and Reasoning of Rotation (importance:15)

Better Retrieval, Limited Clustering Gains: A Controlled Study of Multilingual Company Entity Resolution (importance:15)

DiMOS: Doob-Guided Inference-Time Multi-Objective Search for Scientific Design (importance:15)

Using Evidence-Centered Design to Guide Prompt Engineering in Measurement: An Illustration with Automated Essay Scoring (importance:15)

Understanding Personalization Gains in Whisper ASR (importance:15)

Gradient Boosting Reinforcement Learning (importance:15)

Mitigating Data Reconstruction Attacks in Federated Learning: Conditions, Bounds, and Theoretical Guarantees (importance:15)

The Feasibility of Topic-Based Watermarking on Academic Peer Reviews (importance:15)

Chain of Retrieval: Multi-Aspect Iterative Search Expansion and Post-Order Search Aggregation for Full Paper Retrieval (importance:15)

VeriGRAG: Enhancing LLM-Based Verilog Code Generation with Structure-Aware Soft Prompts (importance:15)

EngTrace: A Symbolic Benchmark for Verifiable Process Evaluation of Engineering Reasoning (importance:15)

Cerium: A Multi-GPU Framework for Terabyte-Scale Encrypted Inference (importance:15)

TTCS: Test-Time Curriculum Synthesis for Self-Evolving (importance:15)

Do We Need Asynchronous SGD? On the Near-Optimality of Synchronous Methods (importance:15)

RoboPocket: Improve Robot Policies Instantly with Your Phone (importance:15)

Muon Dynamics as a Spectral Wasserstein Flow (importance:15)

LLM as a Tool, Not an Agent: Code-Mined Tree Transformations for Neural Architecture Search (importance:15)

Latent Abstraction for Retrieval-Augmented Generation (importance:15)

CyberCane: a privacy-bounded neuro-symbolic retrieval framework for phishing detection under distribution shift (importance:15)

AI estimation of partition functions with the PULSE method for the thermodynamic properties of disordered compounds (importance:15)

GROW$^2$: Grounding Which and Where for Robot Tool Use (importance:15)

GemNav: Discrete-Token Visual Robot Navigation using a Multimodal Large Language Model (importance:15)

How Much AI Is in This Track? Quantifying the Proportion of AI-Generated Stems in Hybrid Music Mixtures (importance:15)

Diagnosing Faults in Reinforcement Learning Simulators with Canonical Polynomial Invariants (importance:15)

LPA-CWM: A Learned Physical Adjudicator for Motion Reasoning with Counterfactual World Models (importance:15)

From Code Archival to Knowledge Graph: Bridging Software Heritage, COAR Notify and Wikidata (importance:15)

AffordanceWAM: Affordance-Aware Joint World-Action Modeling for Robot Manipulation (importance:15)

FedSocket: Recipient-Executable Knowledge Exchange for Heterogeneous Multimodal Federated Learning (importance:15)

MaRK: Markov-adapted Recurrent Kernels for Dynamic Operator Conditioning in State Space Models (importance:15)

MMTA: Benchmarking Multimodal Temporal Analysis with Time Series, Text, and Vision (importance:15)

UniW(A)M: A Unified Mixture-of-Experts Framework for World and World-Action Models (importance:15)

Do Vision CIL Methods Transfer to Protein? A Diagnostic Benchmark of Frozen Protein Encoders (importance:15)

Diagnosing Low-Confidence Patient Profiles via Localized Harmful Covariate Shift Detection (importance:15)

OrionMoE-v1.0: A Mixture-of-Experts Foundation Model for Tabular In-Context Learning (importance:15)

FedPix: Learning What to Share with Less Leakage in Federated Intelligence (importance:15)

QD-Rec: Open-Ended Evolution of Recommendations via Quality-Diversity (importance:15)

Forecasting Patient Trajectories from the Electronic Health Records with Action-Conditioned World Models (importance:15)

SAGE-CP: When Marginal Coverage Hides Route-Conditional Failures in Adaptive VLM Ensembles (importance:15)

Fast Construction of Inner Sphere Trees Using Machine Learning (importance:15)

DS-Net: A Dynamic Split Network for Edge-Cloud Collaborative Inference (importance:15)

SPADE-KD: Peer-to-Peer Collaborative Learning under Heterogeneous Settings (importance:15)

Visual Causal-chain Bookmarking: Retrieval-Weighted Credit Assignment for Reinforcement Learning from Delayed Reward (importance:15)

Looped GNNs Beyond Fixed Depth: Graph Size Generalization and Parameter-Efficient Adaptation (importance:15)

Hidden Trajectories Don't Lie: Correcting Corruption Drift for Test-Time Adaptation (importance:15)

Torch-DPhys: A Torch-Native, MOOSE-Compatible Framework for Differentiable Physics Simulation and Hybrid Machine Learning Integration (importance:15)

Measuring Empowerment from Offline Data for Intrinsically Motivated Exploration (importance:15)

TCMScience: An Autonomous Scientist for Traditional Chinese Medicine (importance:15)

fxBench: Evaluating and Understanding Formula Suggestions in Spreadsheets (importance:15)

A Symmetry-Matching Approach to Blind-Spot Reduction in Sparse Autoencoders (importance:15)

T-MUON: Tensorizing Muon via t-SVD for Large Language Model Pre-training (importance:15)

EasyResearch: Honest Regression, Forecasting and Signal Classification for Researchers Who Do Not Program (importance:15)

Operator-Mismatch Contrastive Learning for World-Relative Trajectory Coherence (importance:15)

Beyond the record: evaluating probabilistic AI weather forecasts deep in the tail (importance:15)

Text Dominance in Paralinguistic Conflict: Tracing Suppression Through Audio LLM Circuits (importance:15)

Curriculum Learning Teaches LLMs Long-Horizon Implicit Reasoning, Up To A Limit (importance:15)

OPDArena: Understanding Learning and Forgetting in On-Policy Distillation (importance:15)

LOOP: Recovery-Aligned Iterative Layer Pruning via On-Policy Distillation (importance:15)

Learning to Couple: Context-Adaptive Joint Denoising for Diffusion Language Models (importance:15)

NADS: Navigator-Guided Data Selection for Mitigating Catastrophic Forgetting in Fine-Tuning (importance:15)

Which Answers Agree? Evaluating Self-Consistency from Incomplete Logs (importance:15)

Asynchronous On-Policy Self-Distillation (importance:15)

FinHopper: A Progressive Multi-Hop Benchmark for Complex Financial Agents with Step-wise Evaluation (importance:15)

A Fisher-Geometric View of Multi-Teacher On-Policy Distillation (importance:15)

ITU-DESC: a missingness-aware machine-learning dementia risk score in the UK Biobank cohort (importance:15)

Topological Sheaves of Irreducibility: A Theoretical Framework for Multi-Agent Task Obstructions (importance:15)

DUET: Decoupled Upper- and Lower-Body Execution for Task-Agnostic Humanoid Loco-Manipulation (importance:15)

BanglaShield: A Script-Robust Safety Classifier and Benchmark for Bangla Native, Romanized, and Code-Mixed Prompts (importance:15)

Who Gets Unmasked? Sampler-Induced Bias in Diffusion Language Models (importance:15)

Measuring Data Diversity with Embeddings: A Taxonomy and Benchmark for Diversity Measures (importance:15)

Estimating Causal Bounds from Observational Data With Neural Generative Models (importance:15)

TILDE: Teacher-in-the-Loop Distillation from Embodied Foundation Models (importance:15)

Clean Forcing: Drift-Resistant Autoregressive Video Diffusion with a Frozen Base (importance:15)

MemT-1.0: Towards Memory-Native Evolution (importance:15)

On the Role of Batch Size in Stochastic Conditional Gradient Methods (importance:15)

AutoREC: A reinforcement learning platform for equivalent circuit model generation (importance:15)

Compute Allocation in Bilinear Contrastive Representation Learning via Scaling Laws (importance:15)

From Failure Taxonomy to Intervention: A Diagnostic Methodology for Industry-Scale AVLM in Video and Live-Streaming Platform Moderation (importance:15)

Systematic Evaluation of Learning Rate Scheduling Strategies Across Heterogeneous Architectures (importance:15)

REATS: LLM Reasoning-based Ensemble Learning for Adaptive Time Series Forecasting (importance:15)

Same path, different walk: a mechanistic comparison of looped and stacked transformer encoders on 12-lead ECG (importance:15)

When Does Backpropagating Through Policy Memory Matter? Separating Representation Credit from Closed-Loop Credit (importance:15)

TRACE: A Reproducible Benchmark for Electricity Price Forecasting with Official Operational Text (importance:15)

Arithmetic Actor Heads and Training Stabilization for Out-of-Distribution Reinforcement Learning (importance:15)

Planetary Geospatial Foundation Models: A New Paradigm for Global Public Health (importance:15)

FreshCast: Timely and Complementary Retrieval for Frozen Time Series Forecasters (importance:15)

AdaptLSTM: Efficient Adaptive Online Learning for Cloud Workload Forecasting under Distribution Drift (importance:15)

Robustly Reliable Learning for Regression (importance:15)

Learning Rational Strategies for Temporal Objectives in Multi-Agent Systems (importance:15)

Residual Stochastic Concept Bottleneck Models (importance:15)

Post-hoc Disentanglement of Textual and Acoustic Features in Pre-trained Speech Encoders (importance:15)

The Computational Complexity of Counting Linear Regions in ReLU Neural Networks (importance:15)

Attention Transfer Is Not Universally Effective for Vision Transformers (importance:15)

Reward and Behavioral Measures of LLM Coordination in Dining Philosophers (importance:15)

REFLEX: Reflective Evolution from LLM Experience (importance:15)

Dissecting Training-Free Uncertainty Estimation in Multimodal Large Language Models (importance:15)

AURAL: Adaptive Latent Reasoning with Joint Chunk for Speech Language Models (importance:15)

Democratizing MoE inference on commodity GPUs with CoMoE (importance:15)

ETH: Data-Balanced Heterogeneous Pipeline Parallelism for Training Large Models (importance:15)

Pattern-Based Precondition Inference from Method Implementations (importance:15)

Step-Aligned Script Generation with Verification and Recovery for Web Automation Testing (importance:15)

Narratives Mislead: An Empirical Study of How LLMs Misjudge CVSS Severity in Bug Bounty Reports (importance:15)

From Evidence to Effect: Authority Semantics and Runtime Infrastructure for Stateful Agents (importance:15)

Making all divisions slower to fix a bug in 0.000000000000000000542% of cases (importance:15)

Rendering untrusted HTML email, safely (2021) (importance:15)

Hanami, Why?: Model Citizens (importance:15)

Opus 5.5 refused to draft/write an example of a paragraph in my thesis (importance:15)

Fable 5.5: what is actually confirmed vs the "tomorrow" rumor (importance:15)

What's one short phrase you keep typing into Claude Code? (importance:15)

I tried to put real craft into a small notch app for working with multiple Claude Code sessions (importance:15)

Disappointed with Cursor. If you switched to Claude Code or Codex, which one? (importance:15)

ChatGPT Coding Environment Suddenly Unavailable in Work Mode. (importance:15)

Improve Qwen3.8-27b low quants by using a prefiller model that streams from SSD (importance:15)

Do not upgrade Strata! Tons of breaking changes! (importance:15)

RunInfra (importance:15)

call4me (importance:15)

Grok Bot just found a pretty satisfying way to annoy the spammers for a change. Its free Spam-Call Scrubber tracks down your phone number on data broker and people-search sites, submits removal requests and follows up on confirmations. Because those lis - X (importance:15)

Stable and time-aware explanations for supply-chain forecasts: A lean XAI framework - Bioengineer.org (importance:15)

SpaceX’s Shotwell: xAI Staff Shift, 0 Tesla Deals [2026] - https://tech-insider.org/ (importance:15)

スピード10倍の開発を支える、AI活用基盤の開発に参戦（インターンレポート） (importance:15)

Adobe Lightroom mobile can now open closed eyes in photos with a simple tap (importance:12)

Podcast: The Future of Web Apps: Will AI Agents Generate UIs on the Fly? (importance:12)

Beyond Explicit Intent: Discovering and Shaping Artistic Intent through Human-AI Co-Creation (importance:12)

MSCR: Jointly Balancing Utilization Across Modalities and Discovering Synergistic Information (importance:12)

CADS: Measuring the Incremental Contribution of LLM Proposers in Sequential Biological Screens (importance:12)

MM-LeJEPA: Correlated Gaussian Regularization for Multimodal Joint-Embedding Predictive Architectures (importance:12)

SQuIREL: A Context-Relative Framework for Measuring Entity Richness in Wikidata (importance:12)

IndustrialCAD-1400: Benchmarking Detailed Industrial Assemblies from Text (importance:12)

Can LLM User-Proxies Stand In for Humans in Reliance Measurement? Replicating Rel-A.I. with Simulated Users (importance:12)

Benefit-Targeted Routing and Bounded Revision for LLM-Based Evidence Screening in Manufacturing Supplier Discovery (importance:12)

TESS: LLM-guided Translation, Exploration, Simulation, and Search for Embodied Task Planning (importance:12)

OrderGPT: Generation Order as a Learnable Modeling Choice (importance:12)

CXBench: A Benchmark for Process-Compliant LLM Agents in Regulated Customer-Experience Workflows (importance:12)

GraphEvoResearcher: Graph-Aware Training-Free Self-Reflective Reasoning for Cost-Bounded Multi-Hop Question Answering (importance:12)

TimesJev: Thinking Fast on Time Series with System 1 Sufficiency (importance:12)

Beyond Static Leaderboards: Capability-Focused Evaluation with Structural Diversity for Time Series Foundation Models (importance:12)

UMI-WM: Interactive World Modeling for Bimanual Manipulation from Wrist Cameras (importance:12)

Zero-Shot Tabular Learning via Hypergraph Reasoning with LLMs (importance:12)

A Group-Aware Comparative Study of Classical Machine Learning and Deep Learning Methods for Sensor-Based Human Activity Recognition (importance:12)

REAP: Learning General Rearrangement Heuristics from Motion-Planner Feedback (importance:12)

PACE: Pareto-Adaptive Compression for Efficient Native MLLMs (importance:12)

TWIN-MoE: Two-stage Intra-Expert Imputation Network for Incomplete Multimodal Learning (importance:12)

Model Capacity Frontier: A Development Framework for LLM-Supported Social Simulation (importance:12)

Bootstrap Stochastic Interpolant Tree for Inference-time Search Under Sparse Reward (importance:12)

Towards an Intelligent Decision Engine for Smart Contract Deployment (importance:12)

Reliability-Preserving Compression for Deployable Medical Image Segmentation (importance:12)

PRIMA: Pose Refined Inductive Motion Adapter for Zero Shot Social Bodily Behavior Recognition (importance:12)

Benchmark for Structured Retrieval-Augmented Generation in Perovskite Photovoltaics (importance:12)

When Better Opportunities Backfire: Service Worker Responses to Parallel AI Deployment in Customer Acquisition (importance:12)

Progress-Guided Preference Generation for Aligning Generative Control Policies (importance:12)

SWIFT: Self-guided Weak-to-strong dIFfusion models via Target extrapolation (importance:12)

PaceVLA: Adaptive Temporal Action Quantization in Vision-Language-Action Models (importance:12)

FrameGRPO: Incentivizing Video Reasoning via Multi-Frame Rollouts and Temporal Credit (importance:12)

TamedFlow: Trajectory-Aware Distillation\\for Teacher-Assisted Diffusion Sampling (importance:12)

Honeyed Words, Hidden Daggers: Benchmarking and Training LLMs to See Through Mock Politeness (importance:12)

SPAR: Structural Pruning and Action Recovery for Vision-Language-Action Models (importance:12)

The Effectiveness and Durability of Pretraining Data Filtering in Protein Language Models (importance:12)

Hierarchical Token Merging with Recursive Autoencoders for Unsupervised Grammar Induction (importance:12)

Game of Phones: Harnessing Game Theory and LLMs in 'Bot Wars' to Counteract Phone Scams (importance:12)

Mise-en-Video: Fine-Grained 3D Control for Multi-Shot Video Generation (importance:12)

AfriBED: a Speech Recognition Bias Evaluation Dataset for Low Resource African Languages (importance:12)

RIFAR: Reliability-Informed and Forgetting-Aware Replay for Continual Robot Learning with World-Action Models (importance:12)

MedZERO: Self-Evolving Agents for Open-Ended Medical Reasoning Through Controlled Knowledge Accumulation (importance:12)

Forecast-PEFT: Parameter-Efficient Fine-Tuning for Pre-trained Motion Forecasting Models (importance:12)

TestDG: Test-time Domain Generalization for Continual Test-time Adaptation (importance:12)

GaussTrap: Stealthy Backdoor Attacks on 3D Gaussian Splatting for Targeted Scene Misperception (importance:12)

Student Perspectives on the Benefits and Risks of AI in Education (importance:12)

To Trust Or Not To Trust Your Vision-Language Model's Prediction (importance:12)

UNO: Unlearning via Orthogonalization in Generative models (importance:12)

NLP Meets the World: Toward Improving Conversations With News Media About Natural Language Processing Research (importance:12)

The Constitutional Controller: Doubt-Calibrated Steering of Compliant Agents (importance:12)

A Privacy-Preserving Federated Learning Method for Heterogeneous Clients (importance:12)

SLA2: Sparse-Linear Attention with Learnable Routing and Quantization-aware Training (importance:12)

Brain-OF: An Omnifunctional Foundation Model for fMRI, EEG and MEG (importance:12)

Bridging the Gap in the Responsible AI Divides (importance:12)

Span-Level Machine Translation Meta-Evaluation (importance:12)

Causal Scene Narration: Inference-Time Text Restructuring for Vision-Language-Action Driving (importance:12)

How a Cooperative-Override Circuit Suppresses Nash Play in Large Language Models (importance:12)

Multilingual Safety Alignment via Self-Distillation (importance:12)

RuPLaR : Efficient Latent Compression of LLM Reasoning Chains with Rule-Based Priors From Multi-Step to One-Step (importance:12)

Parameter misspecification in physics-informed neural networks: A closed-form audit and its limits (importance:12)

Data-Driven Resource Aggregation for Scalable Digital Twin and Simulation Model Generation (importance:12)

The Certification Exponent: Instance-Optimal Sequential Certification of Policy Improvement from History Dependent Logs (importance:12)

Transformers as Adaptive Kernel Learners across Tasks for In-Context Learning (importance:12)

BEM: Learning Bregman Divergences for Concept Erasure in Flow Matching Models (importance:12)

Structured Latent Dynamics for Long-Horizon Embodied World Models (importance:12)

Theory-Guided Instance-Dependent Early Stopping via Second-Order Loss Dynamics (importance:12)

How Does Muon Work in Private Deep Learning? Scalable Differentially Private Training via Orthogonalization (importance:12)

LLM-Based Evaluation Planning for Heterogeneous Optimization (importance:12)

Block Recursive Transformers for Structured Parameter Sharing (importance:12)

Shallow Probes Are Sufficient for Near-Oracle Personalization Depth Allocation in Federated Learning (importance:12)

DROD: Dual-Ratio Ordered Dropout for Federated Continual Learning on Heterogeneous Devices (importance:12)

The Hidden Look-Ahead: Point-in-Time Leakage in Electricity Market Data (importance:12)

Reinforcement learning on the EPRI AI-ccelerating Unit Commitment benchmark: diagnosing over-commitment and guiding MILP with learned policies (importance:12)

FTFM: Factorized Tabular Foundation Models (importance:12)

Finite Interaction Lens: Structured Precision Sharing for Vision-Language Models (importance:12)

Divergence based Ranked Preference Alignment for LLMs (importance:12)

Zephyrus-R1: Agentic Reinforcement Learning for Weather Reporting (importance:12)

How Much Does One Wrong Answer Cost? Error Propagation in Constraint Pruning with Noisy LLM Oracles (importance:12)

Bayesian Model Discovery with Language Models: Closing the Loop between Hypotheses and Experiments (importance:12)

Statistical Learning Theory for Direct Preference Optimization (importance:12)

DeltaTrace: Efficient Signed Attribution for Reasoning Language Models (importance:12)

Truncate or Not? A Scaling Study of Looped Transformers with Truncated Backpropagation (importance:12)

SEAM: Spectral Embedding Alignment and Merging for Low-Resource TTS (importance:12)

Typed Decision Models under Prevalence Shift: Decision Training, Calibration, and Implicit Priors (importance:12)

Global Time-Aware Conditioning (GTC): A Mechanism for Time-Aware Pediatric Sleep Representation Learning (importance:12)

SleepPairNet: Learning Multimodal Pediatric Sleep Representations with Random-Pair Cross-Attention (importance:12)

Deep Pontryagin Learning for Continuous-Time Stochastic Control with Jumps (importance:12)

Cross-Fitting for Neural Posterior Estimation (importance:12)

CAVA: Complementary-Label-Aware Data Valuation (importance:12)

Trajectory Replay: Continuous-Time Mixture Transport for Lifelong Generative Models (importance:12)

Simulation Priors for Rapid Adaptation in Model-Based RL for Autonomous Racing (importance:12)

Training with ML-labeled Data via Stochastic First- and Zeroth-Order Optimization (importance:12)

Learning to Refer: Inference-Time Collaboration in Decentralized Federated Learning (importance:12)

Language-Forcing: Advancing Humanoid Loco-Manipulation via Semantic Alignment of Whole-Body World Action Models (importance:12)

Rethinking weakly supervised localization on chest radiographs: position nulls, protocol dependence, and PACE (importance:12)

Your Self-Supervised Projection Head Captures Object Co-Occurrence Statistics (importance:12)

SPELUNKA: A Framework for Finding Unanticipated Associations in Vision-Language Models (importance:12)

Interpretability of AI weather models via intermediate decoding (importance:12)

Recording-Aware Evaluation of Multimodal OT Intrusion Detection (importance:12)

Training with Ciphers to Enhance In-Context Terminology Learning For Translation from Rare Languages (importance:12)

Federated Learning for Malware Image Classification under Data Heterogeneity (importance:12)

RECAP: Reciprocal Sparse Attention and Hierarchical Caching for Efficient Video Diffusion (importance:12)

Teacher Supervision, Fixed Representations and Calibration: Exact Risk Calculations for Candidate-Scoring Decision Models (importance:12)

Adaptive Dual-Teacher Distillation with Subnetwork Rectification for Bridging Semantic Gaps in Black-Box Domain Adaptation (importance:12)

Post-Selection Distributional Model Evaluation (importance:12)

What Does It Take For A Generalist Robot Policy To Adapt From Minimal Data? (importance:12)

Mask-Aware Execution for Efficient JEPA Training (importance:12)

A Phenomenology of Neural Geometry in Vision-Language-Action Models (importance:12)

When Does UML Software Modeling Pay Off? A Contingency Model of Industrial Modeling Value (importance:12)

Beyond Similarity-Based Generation: Guiding Commit Message Generation with Issue-Commit Traceability (importance:12)

Better Specifications, Better Code: Fine-Tuning LLMs for Behavioral Contract Generation (importance:12)

VerusEval: Evaluating LLM-Generated Verus Specifications Beyond Verifier Acceptance (importance:12)

CAT-RAG: Improving Secure Code Generation with Category-Aware Agentic Retrieval (importance:12)

Which Code Changes Need Re-embedding? CIPBench and the Limits of Multi-Signal Index Maintenance (importance:12)

Updates to site guidelines and AI/LLM tags (importance:12)

Wine ported to OpenBSD (importance:12)

One Hash Bloom Filters (importance:12)

A Very Fast IPv4 Parser, in Swift (importance:12)

A5 v1.0 (importance:12)

What is Haiku 5.5 best suited for? (importance:12)

How do you guys manage the 5-hour limit on Pro without wasting so much usage? (importance:12)

Hitting usage limit fast all of a sudden on Sonnet 5.5 (importance:12)

24GB is plenty if your engines take turns (importance:12)

qev-0.8b · New SOTA tiny local decision model (importance:12)

Hardly any benchmarks for Intel B65, so I got 4x and tested them. Long Context Sweeps for Flash-Next and other models (importance:12)

Agnes-AI/Agnes-3.0-Qwen (QFN finetune) (importance:12)

Trump EPA Can Join xAI Data Center Suit, Fails to Win Dismissal - Bloomberg Law News (importance:10)

Apple broke an essential presentation feature of iPads (importance:10)

NVIDIA Cosmos3-H-Dreams: Bringing Clinical Surgery to Life Through Generative Simulation (importance:10)

Beyond the patch: Protecting your NetScaler gateway against future vulnerabilities (importance:10)

AI adoption by industry shows what other sectors can learn (importance:10)

Plants can speak using electricity, so this startup built an AI model to decipher their language (importance:10)

Discord、Twitch埋め込み経由の1クリックRCEに5500ドルの報奨　force_manifest_nodeからElectronサンドボックス脱出まで連鎖 (importance:10)

Can Compact Open-Weight Large Language Models Recover Mobility Events from Incomplete Trajectories? A Controlled Evaluation Using Air Trips (importance:10)

CME-SpectrumBench: Can LLMs Analyze Condensed Matter Spectral Data? (importance:10)

Continuous Ontology Assurance through External Resources -- An Ontology Engineering Method with a Case Study in Software Language Engineering (importance:10)

A Two-Stage Hybrid Framework for Inferring Air Trips from Mobile Location-based Service (LBS) Data under Incomplete Airport Sightings (importance:10)

Trust the Edge, Doubt the Arrow: Auditing Validation Signals for Text-Mined Clinical Causal Knowledge Graphs (importance:10)

Concordance Plausibility Check for Intervention-Conditioned Human Digital Twins (importance:10)

Frequency-Domain Membership Inference Does Not Beat Confidence Thresholding: A Band-Resolved Analysis of Why (importance:10)

Can AI agents make law? Evaluating AI legal relevance and creativity (importance:10)

From Prediction to Action: Population Health World Model for Digital Medicine (importance:10)

Can LLM Agents Resolve Service Failures? Simulating Multi-Turn Complaint Handling under Information Asymmetry (importance:10)

ReTRACE: Feedback-Guided LLM Agent in a Chemistry Harness for Retrosynthesis Planning (importance:10)

DialogMem: Evaluating Dialogue-Grounded Embodied Memory in 3D Environments (importance:10)

Assessing the AI Readiness of Construction Datasets: A Multidimensional Operational Framework (importance:10)

From Clinical Risk Prediction to Operational Decisions: An Uncertainty-Aware Framework for ICU Staffing (importance:10)

The Forward/Inverse Duality of Controlled Emergence:A Computational Design Program from Explanation to Engineering (importance:10)

BitFlat: Error-Localized Bitstream Design for Robust Diffusion Image Compression (importance:10)

PFedQFT: Personalized Federated Quantum-Inspired Fine-Tuning for Medical Large Language Models (importance:10)

ActTime: Action-Conditioned World Modeling for Time Series (importance:10)

SAR: Learning Global Item-Space Matching and Geometry-Aware Placement Ranking for Buffered Online 3D Bin Packing (importance:10)

EmbodiedWorld-200K: Toward Embodied World Agents via Embodied-Planning Dataset and Interactive World Models (importance:10)

Map-Distribution Mixture-of-Experts for Multi-Agent Pathfinding (importance:10)

FinSafeGuard: A Quality-Driven Synthetic Dataset for Financial AI Safety (importance:10)

CARRS: A Constraint-Aware, Real-Time Routing System for Emergency Vehicles in Urban Environments (importance:10)

Local data augmentation for imbalanced binary classification (importance:10)

Physiology-Aware Graph Attention for Early Ventricular Reentry Prediction (importance:10)

A Semantic Knowledge Graph Construction Pipeline for Vehicle Intention Prediction in nuScenes (importance:10)

Reward Objectives for Robust (PO)MDPs: A Law-Preserving Equivalence with (Partially Observable) Games (importance:10)

LG-Bench: A Graph-Structured Evaluation Benchmark for Life Sciences (importance:10)

Transfer Learning for Power Outage Detection Task with Limited Training Data (importance:10)

Can the Quantum Fisher Effective Dimension Judge a Hybrid Quantum Neural Network? (importance:10)

Composable Parametric Demographic Injection for Values Survey Simulation (importance:10)

GeoForward: Geometry-Guided Cross-View Alignment for Feedforward 3D Gaussian Splatting (importance:10)

SPIRIT: Sparse Physics-Informed Inverse Rendering for SAR Imaging (importance:10)

Improving Fake Profile Detection Through Multimodal Learning and Research-Guided Feature Integration (importance:10)

SECURE-AI: Multi-Signal Uncertainty Quantification for Trustworthy Acoustic Cough Detection (importance:10)

spora: A Unified Multimodal Dataset and Platform for Spatial Proteomics (importance:10)

Client-Aware Security Adaptive Early Exit Vision Transformers for Heterogeneous Federated Learning (importance:10)

On the Selectivity of Generative Models in Structure-Based Drug Design (importance:10)

Landmark-Free and Scale-Blind Multi-View Registration of Preoperative Liver Models to Intraoperative Surfaces (importance:10)

Severity-Channel Conditioning: An Injection-Level Study for Consumer-Device Deblurring (importance:10)

Multimodal-Conditioned Latent Diffusion Models for Spatiotemporal Brain MRI Forecasting (importance:10)

MT-CC: Multi-Group Temperature Scaling for Asymmetric Calibration Behavior in Class-Incremental Learning (importance:10)

OV-PEAR: Observation-Verified Pose-Scale Estimation and Reconstruction for Unknown Industrial Manipulation Objects (importance:10)

Manifold-Constrained Residual Adaptation with Frame-Adaptive Layer Aggregation for Speaker Verification (importance:10)

Not All Similarity Is Leakage: Distribution-Aware Identity Leakage Assessment in Synthetic Face Datasets (importance:10)

GeoEdit: Adversarial De-Evidencing for Silent Geoprivacy Protection via Image Editing (importance:10)

Robotic Photography System based on Vision-Language-Action Model (importance:10)

ALIGN: Aligning Condition Estimation and Expression in Advantage-Conditioned Behavior Cloning (importance:10)

VaDar: Implicit Goal Conditioning via Value Disaggregation (importance:10)

recAMPLIFY: Large-Scale Pretraining of Recursive Protein Language Models (importance:10)

Next Best Action: Causal Constrained Offline RL for Business Action Targeting (importance:10)

FAM-Bench: Benchmarking Condition-Aware Food-as-Medicine Reasoning in Vision-Language Models (importance:10)

CAPT: A Multi-task Continuous Autoregressive Transformer enabling Cross-dataset and Cross-species Transfer for Calcium Population Dynamics (importance:10)

Application of curiosity driven exploration methods for hardware interference identification (importance:10)

Generative Embodied Multiple Behavior Control Systems for Human-like Agents (importance:10)

Modern GLOM: Canonical locks that encode part-whole hierarchies (importance:10)

Controllable GNN Explanations via Multi-Metric Preference Selection (importance:10)

Large Language Models Substantially Compress Well-Being Dispersion but Largely Preserve Its Socioeconomic Structure (importance:10)

DoAtlas-2: A Foundation for Self-Evolving Causal Biomedical Discovery (importance:10)

Just Initialize: A Training-Free Initialization Component for Large-Scale Routing Optimization (importance:10)

FairDiff: Mitigating the Self-Reinforcing Matthew Effect in Diffusion Recommender Models (importance:10)

Calibration-risk routing for controlled world-model adaptation (importance:10)

Evidence Before Sampling: Interpretable Implicit Negative Candidate Discovery for Recommendation (importance:10)

PDB: Not All Drivers Are the Same -- A Personalized Dataset for Understanding Driving Behavior (importance:10)

Dynamic gain neuromodulation attenuates stability gaps during continual joint training (importance:10)

Manifold Approximation leads to Robust Kernel Alignment (importance:10)

Continually Evolving Skill Knowledge in Vision Language Action Model (importance:10)

Patch-Level Tokenization with CNN Encoders and Attention for Improved Transformer Time-Series Forecasting (importance:10)

Machine Unlearning in Low-Dimensional Feature Subspace (importance:10)

Pushing the Frontier of Black-Box LVLM Attacks via Fine-Grained Detail Targeting (importance:10)

What Drives Students' Use of AI Chatbots? Technology Acceptance in Conversational AI (importance:10)

GRAU: A Generic Runtime-Reconfigurable Activation-Requantization Unit for Quantized Neural Network Accelerators (importance:10)

Differential privacy representation geometry for medical image analysis (importance:10)

Rigidity-Aware Geometric Pretraining for Protein Design and Conformational Ensembles (importance:10)

EVA: Aligning Video World Models with Executable Robot Actions via Inverse Dynamics Rewards (importance:10)

OccSim: Multi-kilometer Simulation with Long-horizon Occupancy World Models (importance:10)

ICAT: Incident-Case-Grounded Adaptive Testing for Physical-Risk Prediction in Embodied World Models (importance:10)

A neural operator framework for data-driven discovery of stability and receptivity in physical systems (importance:10)

Forager: a lightweight testbed for continual learning with partial observability in RL (importance:10)

The Causal Description Gap: Information-Theoretic Separations Across Pearl's Hierarchy (importance:10)

BioProVLA-Agent: An Affordable, Protocol-Driven, Vision-Enhanced VLA-Enabled Embodied Multi-Agent System with Closed-Loop-Capable Reasoning for Biological Laboratory Manipulation (importance:10)

Beyond Uniform Local Isometry and Topology: FactoMap for Disentangled Representations (importance:10)

Cleaner Speech, Weaker Generalization: Revisiting Pitt-Derived Benchmarks for Alzheimer's Disease Detection (importance:10)

WILSON - Multi-magnification composites as a foundation for patient-level pathology analysis and diagnostic text generation (importance:10)

Channel-Dependent State Space Model for Multivariate Time Series Forecasting (importance:10)

WeLike2Party! In-Context Motion Transfer for Multi-Human Image Animation (importance:10)

Streaming Multi-Track Timeline Control for 3D Human Motion Generation (importance:10)

Uncovering and Fixing Collider Bias in Bayesian PINNs (importance:10)

GeoBridge: Evolving Transferable Decoder Topologies for Geometric Vision (importance:10)

Pack and Extract: An Effective Graph-Based Approach for Feature Selection (importance:10)

DISCERN: Identifiable Disease Evidence for Generalizable Medical Multimodal Learning (importance:10)

What Remains Without Test Labels? Re-evaluating a Graph-Level Anomaly Detection Benchmark (importance:10)

Dynamic Neural Optimal Transport for Subset Alignment (importance:10)

Power-Law Correlations Without Apparent Temperature-Tuned Criticality in Fixed-Tree Hierarchical Languages (importance:10)

Exact Risk Certificates for Pairwise-Judged Reranking (importance:10)

VBO-MI: A Differentiable Black-Box Optimization Framework via Variational Mutual Information Estimation (importance:10)

HiPPO Functional Memory: Coordinate-Queryable Attention for Irregular Time Series (importance:10)

What "Form'' of Uncertainty Does Conformal Prediction Quantify? (importance:10)

Native Probability Outputs for Clinical Triage: An Evaluation of Calibration and Decision Accuracy (importance:10)

Global Convergence in Deep Networks via the Second Law of Thermodynamics (importance:10)

Distribution-Based Regularization for Reward Fine-Tuning Diffusion and Flow Models (importance:10)

Latent World Models Should Plan with Distances Along the Latent Manifold (importance:10)

Polynomial Learning of Positive ReLU Networks under Gaussian Inputs (importance:10)

FHKAN: Fractional Hermite Kolmogorov-Arnold Networks for Data-Driven and Physics-Informed Learning (importance:10)

Emergent Sparsity: How Many Effective Variables Survive Coarse-Graining in AlphaEarth Embeddings? (importance:10)

A Distribution-Free Separation Between Learned and Fixed Features (importance:10)

Probabilistic Limit Order Book Forecasting with Conditional Diffusion Models (importance:10)

Steered Not Stirred: Effective Reward Tilting using Surrogate Reward Mixing (importance:10)

Adversarial Post-Training of Foundation Models for High-Frequency Trading (importance:10)

Evidence Sufficiency and Model Comparisons in MexRegQA: An Item-Level Audit of Mexican-Spanish Regulatory QA (importance:10)

ACE-Trader: Adversarial Co-Evolution for Generator-Critic Trading Systems (importance:10)

Strength Controls Change How Interventions Compare Across Sparse Autoencoder Codes (importance:10)

GeoSplit: Selective Edge--Server Map Acquisition for Cross-User Travel-Mode Recognition (importance:10)

Doubly Robust Learning of Heterogeneous Survival Treatment Effects under Time-Varying Treatments (importance:10)

Two hidden layer suffice: the caterpillar net (importance:10)

Scaling Parameter-Efficient Fine-Tuning for Brain Encoding (importance:10)

Measuring the Influence of Modality-Specific Node Representations on Graphs (importance:10)

InJePA: Intention-First Latent Planning with Coupled JEPA World Models for Visual Navigation (importance:10)

Vocabulary-Incomplete Unsupervised Vision-Language Model Adaptation (importance:10)

Same Space or Not: Benchmarking Spatial Intelligence Through Cross-View Contradictions (importance:10)

IMPULSE: Push First, Then Predict in V-JEPA 2 Feature Space (importance:10)

Sequence-to-Sequence Neural Surrogate Modeling of Transient Responses in High-Speed Links with Variable Design Parameters (importance:10)

Blackbox Performance Analysis of LLM-based Network Control (importance:10)

Private Evaluation of High-Dimensional Functions (importance:10)

Knowledge Diversion: A Novel Paradigm for Pre-training Decomposable Image Generation Models (importance:10)

LastWords: Boring Problems Revisited (importance:10)

GreenNIDS: A Benchmark Workflow and Analysis Dataset for NetFlow Intrusion Detection Experiments (importance:10)

BiDex-100: A Comprehensive Benchmark for Bimanual Dexterous Manipulation Across 100+ Simulation Tasks (importance:10)

Knowledge Manifold Optimization for Scientific Discovery (importance:10)

Conditionally Identifiable Latent Representation for Multivariate Time Series with Structural Dynamics (importance:10)

Bandit Learning in General Open Multi-agent Systems (importance:10)

ProtSent: Protein Sentence Transformers (importance:10)

BAPR: Bayesian amnesic piecewise-robust reinforcement learning for non-stationary continuous control (importance:10)

Radial-Angular Geometry for Reliable Update Diagnosis in Noisy-Label Learning (importance:10)

Certifying Whether to Adapt: Baseline-Relative PAC-Bayes-MMD Guarantees for Test-Time Adaptation (importance:10)

Encoder-Decoder Manifold Alignment for Idempotent Generation (importance:10)

Dynamic Regime-Aware Conformal Calibration for Reliable Economic Forecast Intervals under Multiple Distribution Shifts (importance:10)

Candidate-Label Saliency for Efficient Conformal Pruning of Text Classifiers (importance:10)

Prior-Amortized In-Context Bayesian Inference for Generalized Linear Mixed-Effects Models (importance:10)

Feature Selection for Tabular Foundation Models: Exploring Feature Utility in Low-Data Molecular Prediction (importance:10)

CTIA-ODE: Continuous-Time Inter-Variable Attention for Long-Term Time Series Forecasting (importance:10)

Dimension-Optimal Dynamic Regret in Exp-Concave Online Learning (importance:10)

Sparse Federated Echo State Networks for Time-Series Classification via Smooth-L1 Regularization (importance:10)

Forecasting Voltage Time Series under Anomalies Using a WGAN Model With Lipschitz Control and Progressive Kernel Critic Strategy (importance:10)

Disentanglement with Biological Constraints: A Theory of Functional Cell Types (importance:10)

Interacting Annealed Importance Sampling via Population-Based MCMC Kernels (importance:10)

Evidence for Phenotype-Driven Disparities in Freezing of Gait Detection and Approaches to Bias Mitigation (importance:10)

Reinforcement Learning for Individual Optimal Policy from Heterogeneous Data (importance:10)

The Super Emotion Dataset (importance:10)

Generative Flows for Potential Outcomes and Counterfactuals (importance:10)

Smooth Flow Matching for Synthesizing Functional Data (importance:10)

Hybrid Topic-Semantic Labeling and Graph Embeddings for Unsupervised Legal Document Clustering (importance:10)

RNAGenScape: Property-Guided, Optimized Generation of mRNA Sequences with Manifold Langevin Dynamics (importance:10)

SparsePixels: Efficient Convolution for Sparse Data on FPGAs (importance:10)

Generative Spatiotemporal Data Augmentation (importance:10)

Statistical Inference and Learning for Shapley Additive Explanations (SHAP) (importance:10)

HyperKKL: Enabling Non-Autonomous State Estimation through Dynamic Weight Conditioning (importance:10)

RARA: Robust Policy Fine-Tuning through Retrieval-Anchored Representation Alignment for Robotic Manipulation (importance:10)

Privacy-Preserving Text Inference for Health-Misinformation Classification and Semantic Retrieval (importance:10)

Lexical Reference Frames: Interpreting LM Residual Streams in Native Semantics (importance:10)

NestedVLA: Learning to Consolidate and Generate Skills for Vision-Language-Action Models (importance:10)

Multi-Agent Strategies for Bridging Programming Language Gaps in Code Generation (importance:10)

LLM-as-a-Judge for Contest-Style Human-AI Co-Creation: A Reliability-Aware Evaluation Framework for Coding (importance:10)

Directed Symbolic Execution for Vulnerability Discovery: An LLM-Guided Approach in KLEE (importance:10)

Combining LLM and Dynamic Analysis for Universal Testing of Static Application Security Testing Tools (importance:10)

The Netherlands is building its own Linux OS to break free from Microsoft (importance:10)

Genera for the Initiated: Inside Symbolics Lisp Machines (importance:10)

That was the best $48.62 spent on Fable. Thank you Anthropic (importance:10)

I used Claude Code to turn 44 real cities into Cities: Skylines 2 zoning maps. 41 players asked for their city on GitHub. (importance:10)

Qwen 3.8 Flash Next: drastically faster and much more precise in reaching the result. (importance:10)

Kling 4.0 (importance:10)

Tiny Little Office (importance:10)

OpenShip Cloud (importance:10)

Lugh Xtudio (importance:10)

Asimov 1 (importance:10)

Latest open artifacts (#25): DeepSeek V4.1 Flash 1017, American-made models, and new regulatory controls in the US & China - Interconnects AI (importance:10)

Scientists say Elon Musk is getting something wrong about whether Grok is getting tortured - Business Insider (importance:10)

メルカリにおけるTiKVオートスケールの実践 (importance:10)

MiniMax, DeepSeek Unveil Faster, Cheaper AI Models - 조선일보 (importance:8)

EMudhra Intermediate CA Compromise (importance:8)

How I got over 300 Tor exit nodes removed (importance:8)

JetBrains .NET Day Online 2026: Watch the Recordings (importance:8)

Edge AI for Smart Cities: From Insight to Cognitive Action (importance:8)

$M^2Net$: Unifying Scale-Specific Structures and Prototype Decomposition for Spatial-Temporal Forecasting (importance:8)

Rethinking EEG Spatial Super-Resolution as Dynamic State Decomposition and Filtering (importance:8)

Spherical Physics-informed Neural Operator with Multi-scale Coupling for Meteorological Downscaling (importance:8)

HyperTaxa: Multiscale Genome Representations for Generalized Microbial Taxonomic Inference (importance:8)

TRACE: Tracking Multimodal Clinical Trajectories for Inpatient Risk Prediction (importance:8)

When to Stop Asking: Factor Coverage Bias and LLM Determinability Judgments with Underspecified Information in Medicine (importance:8)

Minimum-Cost Robust Evidence Certificates for Horn Reasoning under Correlated Source Failures (importance:8)

CODA: Cohort- and Drift-aware Foundation Model for Multimodal Clinical Reasoning (importance:8)

Motor compound fault diagnosis using an attention-based cross-talk architecture (importance:8)

Global probabilistic forecasting of air pollution (importance:8)

MoveBench: A Rubric-Based Benchmark for Evaluating Large Language Models in Sports and Health Coaching (importance:8)

Fairness or Fallacy? Mapping the Landscape of AI Bias in Disease Detection (importance:8)

Towards More Granular Reasoning: Mapping the DSA Transparency Database Taxonomy to Platform Terms and Conditions Using Large Language Models (importance:8)

CoRS: Fast Neural Flexible Job-Shop Scheduling via Composite Interval State Reuse (importance:8)

Machina Pauliana: A neuro-symbolic approach to the actio pauliana in Dutch insolvency law (importance:8)

LOCU: Lowdin-Orthogonalized Constraint Updates for Multi-Constraint Policy Optimization (importance:8)

Transformer-based segmentation of prosodic boundaries in Brazilian Portuguese (importance:8)

Context-Aware RGB-Thermal-Radar Obstacle Perception for Resource-Constrained ASVs (importance:8)

CrowdH-Edit: Reference-Guided Instruction-Based Synthetic Hajj Crowd Image Generation for Crowd Counting (importance:8)

SituHOI: Scalable Capture of Scene-Situated Full-Body Human-Object Interaction (importance:8)

IWOO: Wave Optics Optimizer Improved with Hybrid Strategies for Numerical Optimization and Engineering Applications (importance:8)

BLANC: Adaptive Temporal Scheduling for Sign Language Production (importance:8)

Graph-Regularized Synthetic Control for Stable Counterfactual Estimation under Donor Redundancy (importance:8)

Residential Construction Image Dataset Using CSI Divisions for Deep Learning-based Object Detection (importance:8)

ACorD: Adaptive Correspondence Discovery for Medical Image Registration (importance:8)

Corrected Block Sparse Spectral Optimization: Exact Models, Finite Identification, and Local Dynamics (importance:8)

AirMPA: A Meteorology-to-Pollution Adapter for Global Air Quality Forecasting (importance:8)

Optimizing Marketing Budgets for Social Influence: A Generalized Fractional Incentives Framework (importance:8)

Multimodal Decoupling for Diversified Recommendation through Semantic and Structural Shaping (importance:8)

Toward Online Robust Zero-Sum Markov Games with Function Approximation (importance:8)

CIS-Transolver: A Cross-Layer Incremental-State Neural Operator for Fine-Scale Automotive Aerodynamic Flow Prediction (importance:8)

PhyPro-FM: A Physics and Prompt-Guided Graph Foundation Model for Contingency Analysis (importance:8)

Learn the Task, Not the Schedule: Reducing Spurious Cross-Arm Coupling in Bimanual Imitation Learning (importance:8)

Conducting Reliable Keyword Estimation with Unreliable ASR in Police Audio Data (importance:8)

CoopUQ: Robust Multi-Robot Social Navigation via Uncertainty-Aware Cooperative Forecasting (importance:8)

Completion-Aware Cross-Fidelity Offline-to-Online Reinforcement Learning for Multi-Line Bus Holding (importance:8)

Data-driven multi-agent modelling of calcium interactions in cell culture: PINN vs Regularized Least-squares (importance:8)

Equivariant Volumetric Grasping (importance:8)

SlotMatch: Distilling Object-Centric Representations for Unsupervised Video Segmentation (importance:8)

Skin Tokens: A Learned Compact Representation for Unified Autoregressive Rigging (importance:8)

SF-Mamba: Rethinking State Space Model for Vision (importance:8)

Derived Fields Increase the Recoverable Information in Budgeted Neural PDE Surrogate (importance:8)

Bringing Up a Bilingual BabyLM: Investigating Multilingual Language Acquisition Using Small-Scale Models (importance:8)

From Scientific Crop Imagery to Agronomic Reasoning: A Multimodal LLM Benchmark for Plant Phenotyping (importance:8)

End-to-End Radar and Communication Modulation Recognition with Neuromorphic Computing (importance:8)

Announced Breaks Separate Conformal Reliability from Frequency Calibration (importance:8)

Physics-Aware Self-Supervised Pretraining on Multi-Domain Raman Spectra (importance:8)

GLOW: General Linear Models Optimized with Ward's Method (importance:8)

Budgeted Bayesian Bandits for Future Utilization (importance:8)

Margin Certificates for Voting Classifiers under Mixing Dependence (importance:8)

Latent State Utilization in Probabilistic Circuits (importance:8)

PAC Learning with Query Feedback (importance:8)

Uniform convergence at optimal rate and scale (importance:8)

Revisiting Smoothed Online Learning: Competitive Ratio, Dynamic Regret with Switching Cost, and Best of Both Worlds (importance:8)

Cross-Dimension and Cross-Modality Relational Tensor Representation Learning for Multimodal Health Data Modeling (importance:8)

Transport What Shares the Conduction Pathway: Normal-Anchored SPD Geometric Transport for Source-Free Patient Adaptation of Holter and Patch ECG (importance:8)

Near-Optimal Regret for Semi-Bandit Learning of Monotone Stochastic Problems (importance:8)

Generative Retrieval for Reverse Target Fishing with a Proteome-Scale Benchmark (importance:8)

TopoCompile: Molecular Generation with Exact Global Count Constraints (importance:8)

From Score to Diagnosis: Search-Independent Post-Diagnosis for Symbolic Regression (importance:8)

A Multi-Step Inertial ADMM Algorithm for Federated Learning (importance:8)

Representations Have Trajectories: Characterizing Temporal Representation Dynamics in Neural Networks (importance:8)

DeliMF: Gated Cross-Attention Fusion of Structured Time Series and Clinical Text for Early Intensive Care Delirium Prediction (importance:8)

Uncovering Semantic Hierarchies in Text-Attributed Graphs via Variational EM-based LLM-GNN Synergy (importance:8)

Gaussian Process Bandits with Paired Auxiliary Observations (importance:8)

From Overparameterised Expressions to Parsimonious Models: Fisher-Geometric Pruning for Symbolic Regression (importance:8)

Paired-Outcome Learning for Selective Spacecraft Attitude Capture under Torque Limits (importance:8)

Process-Aware Event Log Augmentation for Process Prediction under Labeled Data Scarcity (importance:8)

Compose the Horizon: Parameter-Shared Blockwise Forecasting under a Sequential-Compute Trade-off (importance:8)

An Impossibility Result for ETF Target Geometry, and an Equinorm Remedy for Imbalanced Analytic Continual Learning (importance:8)

Sample-Efficient Reinforcement Learning with General Utilities via Neural Dual Critics (importance:8)

CARTA: Self-Supervised Reusable Transaction Embeddings via Heterogeneous Tokenization and Cross-Attention (importance:8)

Multifidelity Monte Carlo Estimator for High-Dimensional Quantities of Interest (importance:8)

Sampling and Curvature in First-Exit Comparisons for Momentum SGD (importance:8)

Data-driven forecasting of tropical cloud forest phenology: A dynamical systems and machine learning hybrid approach (importance:8)

A discrete-continuous-time stochastic framework for attention-deficit/hyperactivity disorder-related symptom modeling and pharmacological intervention via methylphenidate (importance:8)

Predicted Class Correction: Conformal Thresholds for Classes with No Calibration Data (importance:8)

Multi-site Learning for Forecasting Day-Ahead Wind-Power Production for Onshore Wind Farms in Great Britain with Deep Neural Networks (importance:8)

Simulating Infant Senses with Age-Tuned Networks (importance:8)

Structured Relevance Determination and Channel Additive Gaussian Processes for Image Data (importance:8)

Paloma: Phase-Conditioned Residual Modulation for Time Series Forecasting (importance:8)

Near-Optimal Last-Iterate Convergence of SGD in the Smooth Interpolation Regime (importance:8)

KPI-Aware Multimodal Anomaly Detection for 5G and Open RAN Telemetry (importance:8)

FracTS: Hierarchical and Autoregressive Time Series Generation (importance:8)

Predictive Bandit Curriculum Reinforcement Learning via Gradient-Based Sample Group Discovery (importance:8)

Neural Video Coding under Limited Delay: Theoretical Optimality and System Design (importance:8)

Graph-to-Table Interfaces Control Relational Generalisation (importance:8)

From Retrieval to Recognition:Knowledge-Enhanced Post-training of Time Series Foundation Models for Classification (importance:8)

Trust Your Targets: Is Target Propagation Block-Diagonal Gauss-Newton? (importance:8)

Learning SOCP Solver Dynamics with Graph Neural Networks (importance:8)

Logit Averaging During Training Sets a New State of the Art for Non-Invasive Speech BCI (importance:8)

Wet-Biased Data Pipelines Mislead the Training and Evaluation of Generative Precipitation Nowcasting (importance:8)

EAGLE: Entropy-Aware Geometric Learning Enhancer for Partially Polygon-Supervised Object Localization (importance:8)

Geo-vision: Native-Resolution Geometry and Sampled Topology for Vehicle-Part Damage Classification (importance:8)

Spatio-Temporal Detection of Photovoltaic Systems on Industrial Building Rooftops Using Satellite Imagery and Machine Learning (importance:8)

Super-resolution as Spatial Downscaling: Coherence, Structure and Task Utility in Earth Observation (importance:8)

Weakly Supervised Discovery of Motion Styles for Compositional Robot Control (importance:8)

Reducing double-dipping in single-cell topic models with three-way count splitting (importance:8)

Candidate-dependent residual routing trades top-of-list accuracy for recall in sequential recommendation (importance:8)

Censored Flow Matching for Synthetic Time-to-Event Generation (importance:8)

CoCoNet: Convex Conformal Network for Multi-Dimensional Uncertainty Regions (importance:8)

Learning from the Right Regions: Class-Conditioned Explanation Guidance for Medical Imaging (importance:8)

Denoising Implicit Variational Inference (importance:8)

Reconstruction of Unobserved Core Electron Density in Tokamak Plasmas with a Neural Operator Transport Surrogate and Uncertainty Quantification (importance:8)

Conditional Weight of Evidence for Gradient-Boosted Credit Models: Estimation, Additivity and Its Relation to Adverse Action Reasons (importance:8)

Lightweight Machine Learning for Classifying Disaster-Related Social Media Posts (importance:8)

Query-Dependent Reading for Quantized Inner-Product Estimation: A Minimax Separation (importance:8)

No Escape from the Quadratic Barrier: Adaptive Attacks on Linear $\ell_2$ Sketches (importance:8)

RAIL-5G: A Challenging Real-World Dataset and Benchmark for AI-Based 5G Indoor Localization (importance:8)

LiftDiff: Guided Discrete Diffusion for Quantum Code Construction (importance:8)

Fair Regression under Approximate Demographic Parity (importance:8)

FLMNet: Self-Ensemble Distillation from Masks to Seeds for Single-Tower CTR Prediction (importance:8)

ForagerSim: Evaluating Embodied Map-Based Reasoning in Large-Scale Dynamic Worlds (importance:8)

Gradient descent reaches arbitrary accuracy in a fixed number of large steps on multiclass logistic regression (importance:8)

Cooperative Multi-Agent Reinforcement Learning for Market Making on Prediction-Market Order Books (importance:8)

D-STEER: Two-Way Control of DPO-Induced Safety Behavior in Activation Space (importance:8)

A Transport Geometry of Benign Overfitting (importance:8)

Best of both worlds: Stochastic & adversarial best-arm identification (importance:8)

Integrative neurocybernetic modeling in the era of large-scale neuroscience (importance:8)

Optimizing Social Utility in Sequential Experiments (importance:8)

ForcingDAS: Unified and Robust Data Assimilation via Diffusion Forcing (importance:8)

Buffer-Parameterized Machine Learning Surrogate Models for Cross-Technology Signal Integrity Analysis and Optimization (importance:8)

FracEvent: Event-Camera Simulation via Fractional-Relaxation Pixel Dynamics (importance:8)

Machine Can Automatically Discover Parametric Functions to Model HEP Data (importance:8)

Propensity Straight-Through Gradients for Discrete Stochastic Systems (importance:8)

Deep Generative Crystal Structure Prediction: A Benchmark Study and a Controlled Test of Prototype Dependence (importance:8)

MADGRAV: a multilevel anomaly-detection pipeline for gravitational-wave searches applied to LIGO data (importance:8)

How the Classical Baseline Shapes Machine-Learning Speed Claims in Topology Optimization (importance:8)

Safety Analysis of CPSs Over-the-air Updates: Test Case Prioritization and Selection (importance:8)

Effective Construction of Feature Database of Fingerprinting Functions for Code-Clone SCA (importance:8)

RXSynth: A Seed-Grounded Corpus for Studying Java Runtime-Exception Bugs (importance:8)

Theory of Developer Momentum: The Developer's Experience of Progress and the Craft of Maintaining Motion (importance:8)

What Is Under the Hood? A Cross-Ecosystem Study of Vulnerable Dependencies in Construction Software (importance:8)

Function Contracts for Bounded Model Checking of C Programs (importance:8)

Backend-Frontend Converter: Component-Based Conversion of .NET Computational Models to TypeScript (importance:8)

Parameter-Efficient Multi-Task Fine-Tuning in Code-Related Tasks (importance:8)

Instrumentation Optimization for Practical Dynamic Race Detection (importance:8)

Protecting CPU AI On Edge TEEs: WebAssembly's Promise and Practical Challenges (importance:8)

DICE: Tilting the Odds through Interface Attacks on System One Model (importance:8)

Evaluating and Improving Static Application Security Testing Tools for Rust (importance:8)

The Annotated VT100 Firmware (2022) (importance:8)

Attended my first Claude hackathon and learned that building a good product and winning a hackathon are two very different things (importance:8)

Old TCL Beyond TV was super slow, so I tried this open-source debloat with Claude and it worked (importance:8)

Which old "hardware" did you "revive" with claude? (importance:8)

Why is Claude taking up SO MUCH space? (importance:8)

Worked the past months to make a point-and-click adventure game engine with Claude (importance:8)

How do I make AI actually think and execute instead of constantly babysitting it? (importance:8)

Real Engineers Don't Use AI (importance:8)

MiMo 2.6 is reasoning is like the crash test dummies (importance:8)

[AINews] not much happened today (importance:8)

Sorcerer (importance:8)

Stage (importance:8)

xtty (importance:8)

Point and Yap (importance:8)

LowNet: Looped Wavelet Neural Networks with 24 Learnable Parameters for Solving Poisson Equations (importance:6)

CTFS : Collaborative Teacher Framework for Forward-Looking Sonar Image Semantic Segmentation with Extremely Limited Labels (importance:6)

Estimating Individual Tree Height and Species from UAV Imagery (importance:6)

Voronoi-guided Bilateral 2D Gaussian Splatting for Arbitrary-Scale Hyperspectral Image Super-Resolution (importance:6)

Diversity-Aware Active Learning Builds Materials Datasets for Current and Future Objectives (importance:6)

Decentralized Online Exp-Concave Optimization (importance:6)

The Optimal Sample Complexity of Pointwise Replicable Agnostic Learning (importance:6)

CVTA: Cross-Variable Temporal Attention for Multivariate Irregular Time Series Prediction (importance:6)

Distributionally-Robust Policy Learning with Observational Data (importance:6)

PRISM: Prior Relational Information for Self-supervised Modeling to Enhance Solubility OOD Generalization (importance:6)

Communication-Efficient Federated Domain Generalization in Edge Networks via Dynamic Prototype Regulation (importance:6)

Out-of-Distribution Detection in Continual Learning (importance:6)

Provable Adversarial Detection: Prime Quantization meets Gromov--Wasserstein (importance:6)

Dynamical Batching: How Class-Cycling Training Shapes the Internal Structure of Neural Networks (importance:6)

CAST: Cast-Admissible Generative Design and Shape Optimization with Paired Depth Images (importance:6)

Certified neural approximations for parabolic obstacle problems (importance:6)

Extending Myerson's Optimal Auctions to Correlated Bidders via Neural Network Interpolation (importance:6)

Modeling Uncertainty Propagation in 3D Morphable Models (importance:6)

Geometry-Aware Interpolation for Inverse Problems (importance:6)

Is User Calibration Still Necessary for Continuous Myoelectric Control? A Closed-Loop Comparison of Calibration-Free Controllers (importance:6)

Tilted Mondrian: Triage and Mitigation of Fine-grained Disparities in Conformal Prediction (importance:6)

DF-PALM: Dynamic Fusion Prompt Learning for Audio--Language Models (importance:6)

Integrable building blocks for symplectic neural networks: learned-type layers, fiber-preserving layers, and what a single layer can do (importance:6)

Decision-Target Reversal under Informative Label Missingness (importance:6)

HullProbe: Measuring Interventional Part Reliance in Maritime Vessel Classifiers (importance:6)

Bayesian Preference Learning for Test-Time Steerable Reward Models (importance:6)

In Context (importance:6)

Ofcom concedes defeat in 4chan case (importance:5)

NixOS is making me less afraid (importance:5)

MicroG: Free-libre re-implementation of Google's proprietary Android user space (importance:5)

Called to serve: Tech, research, and positive impact (importance:5)

Small Talk with Ulises Olave, CTO and Co-Founder of Fitia (importance:5)

On the Benefits of Uncertainty Estimation for Species Classification: A Case Study on Optimizing Expert re-Annotation (importance:5)

On Counting Graphs for Modal Logic with Counting (importance:5)

REFLEX: RNA Ensemble Generation via Flexibility-Calibrated Stochastic Bridge (importance:5)

Analysis of Binary O-type Star Systems Using Recurrent Neural Networks (importance:5)

Reconstruction and Discrete-Event Simulation Assessment of an Analytical (importance:5)

How Robust Are Prognostic Radiomic Features? Comparing Selection Techniques on Pooled Glioblastoma Cohorts (importance:5)

Physics-anchored residual quantum neural networks for cloud-cover parameterization (importance:5)

MCDA: Multi-View Graph Contrastive Learning with Dynamic Topology Augmentation for Functional MRI Analysis (importance:5)

Python Tool for Visualizing Variability of Pareto Fronts over Multiple Runs (importance:5)

Differential Functionalism: Jacobians Shape the Structure of Phenomenal Experience (importance:5)

An integrated interpretable control effectiveness learning and nonlinear control allocation methodology for overactuated aircraft (importance:5)

Neptuna: A Comprehensive Machine Learning Framework for Benchmarking Complex Multiphase Flows (importance:5)

Beyond density snapshots: Dynamical memory in Chern-Simons reservoirs (importance:5)

BridgeCast: Bridging Ocean Wave Forecasts to Reanalysis via Flow Matching with Exogenous Variables (importance:5)

Deep Learning-based Semantic Segmentation of Dental Radiographs: Comparative Evaluation of PSPNet, DeepLabv3, UNet, and UNet++ (importance:5)

Inducing Fourier Geometry: A Smoothness Prior Imposes Approximate Translation Equivariance in Multimodal Stellar Embeddings (importance:5)

Constructive Universal Approximation by Infinite-Dimensional Neural Networks via Gaussian-Weighted Ridgelet Transform (importance:5)

No-Regret Revenue Maximization in Sequential Market Making Under Inventory Constraints (importance:5)

Recurrent Temporal Restricted Boltzmann Machines as Unsupervised Encoders for Biomechanical Movement Sequences (importance:5)

Semi-supervised learning on singular data sets (importance:5)

Conditional Average Treatment Effect Distillation for Hypothetical State-Intervention Targeting Under Budget Constraints (importance:5)

Can Metadata Fix the Gauge? Calibration Turns Sparse Multi-Source Learning from Sparse PCA into Sparse Mean Recovery (importance:5)

Stochastic Tracking of Drifting Affine Fixed Points (importance:5)

Detection Augmented Learning In Non-Stationary Bandit Environments (importance:5)

Sector-Ring Linear Reservoirs: Sector-Wise Decay-Timescale Allocation With State Reuse Across Candidate Settings (importance:5)

Residual-Guided Adaptive Tucker Rank Allocation for Parameter-Efficient MCI Conversion Prediction (importance:5)

An Analysis of Semantic and Spatio-Temporal Co-evolution for Edge Classification on Dynamic Text-Attributed Graphs (importance:5)

PnP-CF: A Consensus Flow-Matching Solver for General Inverse Problems (importance:5)

Scenario Generalization and Sector Transfer in Real Surface Irrigation Surrogate Modelling with Tabular Models (importance:5)

ProtGlycanDock: Towards Accurate Protein-Glycan Docking with Tailored Dataset, Benchmark and Model (importance:5)

Learning from a Bandit Learner (importance:5)

Declarative Explanations for Graph Neural Networks: A Demonstration (importance:5)

Variance-Aware Non-Stationary Bandits:All-Budget Regret Bounds and Adaptation Costs (importance:5)

Optimal Multi-Kernel Combination for Online Multi-Label Classification (importance:5)

TaQFormer: A Treatment-as-Query Transformer with Data-Fusion for Multidimensional Treatment Modeling in Coupon Allocation (importance:5)

OceanTensorCube and ST-MMT: A Multimodal Data and Transformer System for Seven-Day Forecasting of Pyropia Bleaching (importance:5)

CTLite: Lightweight Irregular Multivariate Time Series Forecasting with Reusable Continuous-Time Summaries (importance:5)

A Simple and General Approach to top-$k$ Selection (importance:5)

On Learning Expand-and-Sparsify Representations (importance:5)

A Simple and Unified Method for Gradient-Variation Regret (importance:5)

OrTeN: Neural Inverse Graphics for Joint Terrain and Line-scan Camera Trajectory Reconstruction (importance:5)

A decomposition-integration framework coupling singular spectrum analysis with a volatility-adaptive LSTM for stock price forecasting (importance:5)

Learning volatility-adaptive forgetting via GARCH-modulated LSTM for stock price forecasting (importance:5)

Prototype-Guided Optimization of Quantum Data Embeddings with Linear Cost (importance:5)

Margin-Based Generalization for Quantum Machine Learning with Classical Data Encoding (importance:5)

Accelerating Finite-Sum Bilevel Optimization with Fully First-Order Oracles (importance:5)

Martingale Posterior Cox Processes (importance:5)

Scalable and Adaptive Distribution Regression via Random Features (importance:5)

ATLAS: Expert-Guided Sustainable Satellite Backhaul for Off-Grid Wildlife Monitoring (importance:5)

Normalizing Flow for Neural-Enhanced Track-Before-Detect (importance:5)

Non-Acceleration of the Heavy-Ball Method in 1D (importance:5)

ChronoChart: Setup-Agnostic Channel Charting for Label-Free Indoor Localization (importance:5)

Whistle and Wrong Lyrics: Synthetic Queries Are Not Enough for Real-World Query-by-Melody (importance:5)

CheXFuse: Guided Differential Joint Attention Fusion for Robust Multimodal Chest X-Ray Classification in Data-Scarce Medical Settings (importance:5)

Fundamental Limits of Information-Theoretic Secure Aggregation with User Dropouts in Federated Learning (importance:5)

TriScore: Post-Hoc Out-of-Distribution Detection with Energy, Boundary Probes, and Transform Consistency (importance:5)

Selecting the Best Few with Human Evaluators: Adaptively Combining Ratings and Pairwise Comparisons (importance:5)

HyPE: Attention with Hyperbolic Biases for Relative Positional Encoding (importance:5)

Federated Learning: From Theory to Practice (importance:5)

Optimizing Prompt Sequences using Monte Carlo Tree Search for LLM-Based Optimization (importance:5)

AutoTailor: Automatic and Efficient Adaptive Model Deployment for Diverse Edge Devices (importance:5)

TimeART: Towards Agentic Time Series Reasoning via Tool-Augmentation (importance:5)

Sparse Subspace-to-Expert Sharing for Task-Agnostic Continual Learning (importance:5)

Nearly Optimal Algorithms for Adversarial Kernelized Bandits (importance:5)

Approximation Theory of Laplacian-Based Neural Operators for Reaction-Diffusion System (importance:5)

Microstructure-Conditioned Surrogate Models for Graded Multiscale Optimization of Mycelium Composites (importance:5)

Task-Oriented Sensing and Covert Transmissions for Collaborative Multi-AUV Systems (importance:5)

Hierarchical Multilevel Monte Carlo for Order-Optimal Neural Actor-Critic in Average-Reward CMDPs (importance:5)

Capacity Confounds and Coverage Guarantees in Adaptive Sub-model Federated Learning (importance:5)

When Clean Data Hurts: Learning with Monotone Corruptions Beyond Binary Classification (importance:5)

Characterizing Client-Independent Stationary Bias for SCAFFOLD and FedAvg in Federated Learning (importance:5)

Spectral Features Dominate BCG Respiratory-Event Detection: A Large-Scale Patient-Independent Comparison of Feature Groups in Sleep Apnea Patients (importance:5)

Trace-Norm Geometry of Composed Multitask Predictors (importance:5)

Transferable Mass Spectrum Prediction via Reference-Guided Test-time Specialization (importance:5)

Stochastic Optimization Under Power-Law Spectra: Tight Bounds and Shuffling Analysis (importance:5)

Spacing Test for Fused Lasso (importance:5)

Geodesics in the Deep Linear Network (importance:5)

Refined Gradient-Based Temperature Optimization for the Replica-Exchange Monte-Carlo Method (importance:5)

Breaking the Stochasticity Barrier: An Adaptive Variance-Reduced Method for Variational Inequalities (importance:5)

Control Variate Design for Score-Function Gradient Estimators in Importance-Weighted Variational Inference (importance:5)

A Unified Physics-Informed Neural Network for Modeling Coupled Electro- and Elastodynamic Wave Propagation Using Three-Stage Loss Optimization (importance:5)

Data-Efficient Indentation Size Effect Mitigation in Steels Using Machine Learning and a Physics-Constrained Neural Network (importance:5)

PACE: Geometry-Aware Bridge Transport for Single-Cell Trajectory Inference (importance:5)

Machine Learning for Coding Retail Product Names to Consumer-Price Categories: A Rule-plus-Bag-of-Words Pipeline with Reliability-Weighted Human-in-the-Loop Labeling (importance:5)

Regional summaries of Bayesian model uncertainty (importance:5)

FairBED: Incorporating Fairness into Bayesian Experimental Design (importance:5)

Tropical Circuits with Scalar Multiplication Gates (importance:5)

Gravitational-wave parameter estimation with machine-learning generated surrogate waveforms (importance:5)

Harmonic Torsional Diffusion for Protein-Ligand Flexible Docking (importance:5)

Advanced Linear Algebra with Applications - Part I (Numerical linear algebra for PDEs, machine learning, and data assimilation) (importance:5)

Arbitrage-Aware Probabilistic Multi-Step Forecasting of Implied Volatility Surfaces: Joint Generation of Surface-Return Trajectories Using Latent Diffusion (importance:5)

Parameterized Complexity of $L_p$-Lipschitz Constants for Input Convex Neural Networks and $L_p$-Norm Maximization over Zonotopes (importance:5)

A Splitting Method for SDE Terminal-Law Estimation (importance:5)

LandscapeSHAP: Which Persistent Homology Class Gets the Credit? (importance:5)

Registration is a Representation Problem: Solving Whole-Body PET/CT Registration (importance:5)

Distinguishing Process Change, Instrument Fault and Data-Path Corruption in Streaming Telemetry: A Reference Implementation and Benchmark (importance:5)

Noise-Awareness in Quantum Software Testing: An Empirical Landscape Analysis (importance:5)

The EBC Proportionality Law: A Recurring Architectural-Role Pattern Across Java Software (importance:5)

A GNN-Based Approach for Code Smell Detection and Refactoring Opportunity Identification (importance:5)

NosRacer: Dynamic Detection of Race Conditions in On-Device Network Operating Systems (importance:5)

OpenBSD 8.0 (importance:5)

Rhapsody in Clawd (importance:5)

I’m happy with Opus 5.5. Stop everything, we don’t need more (importance:5)

How to get the most out of Claude Pro to maximize a job search? (importance:5)

Since TV debloating seems popular, here's mine: an open-source replacement homescreen + debloater for Fire TV (no root required) (importance:5)

Play around with Jev as a chatbot (importance:5)

[Bi-Weekly Megathread] Project Showcase (importance:5)

Short report from Lithuanian conference on AI Technologies in Medicine (importance:5)

chinese hardware (importance:5)

/dev/null (importance:5)

reception (importance:5)

croft (importance:5)

Rusteerio by Context.dev (importance:5)

AttributedString構造で実現した大規模なiOS設定システム (importance:5)

Federated Community Detection in Bipartite Networks (importance:4)

Towards instance-dependent regret optimality in Episodic MDPs with Posterior Sampling (importance:4)

Multilingual Voicemail and Answering Machine Detection via Pretrained Audio Embeddings and Gradient Boosting Machines (importance:4)

A Pitfall in the Evaluation of Gene Perturbation Models (importance:4)

LaMER: Language-Conditioned Multi-Expert Routing for Weakly Supervised Referring Expression Comprehension (importance:4)

Recurrent Self-Attention Dynamics: An Energy-Agnostic Perspective from Jacobians (importance:4)

Robust and Efficient Finetuning of Vision Foundation Models via Implicit Ensembling (importance:4)

A Unified Framework for Zero-Shot Reinforcement Learning (importance:4)

Standardization of Post-Publication Code Verification is Possible with the Support of the Community (importance:4)

Quantifying Memorization and Privacy Risks in Genomic Language Models (importance:4)

The fossil fuel comeback is here (importance:3)

The edge-based contiguous p-median problem: exact formulations and strengthening inequalities (importance:3)

Buy the Rumor, Sell the News: When Is News Priced In? (importance:3)

Joint Deployment and Resource Management for Group-Tag-Based Physical-Layer Authentication and Collaborative Communication in NTN (importance:3)

Breaking the 1/3 Barrier for Approximate Nash Equilibria in Symmetric Bimatrix Games (importance:3)

Concurrent two-scale topology optimization of printable bone scaffolds via rectified-flow generation and exact latent-space refinement (importance:3)

Uncertainty-Gated SHAP-Dirichlet Refinement for MRI-Based Depression Classification (importance:3)

Arbitrage-Free Implied Volatility Surface Reconstruction via Additive Martingales (importance:3)

Reinforcement Learning Controllers for Fusion Scenario Development (importance:3)

Physics-Driven Framework for QoS-Constrained Satellite System with Non-Linear PA Distortion (importance:3)

SafeWall: A Risk-Aware Job Walltime Recommendation for HPC Scheduling (importance:3)

Injection of Space Radiation Errors on Low-power NVM using Generative Deep Learning (importance:3)

Supervised Learning with Quantum Hamiltonian Embeddings (importance:3)

Newton-Based Methods for Bilevel Optimization Beyond Lower-Level Strong Convexity (importance:3)

A probability calibrator for machine-learning classifiers (importance:3)

Chaos persists in large-scale multi-agent learning despite adaptive learning rates (importance:3)

Provable Bounds on the Hessian of Neural Networks: Derivative-Preserving Reachability Analysis (importance:3)

Heterogeneous Sheaf Neural Networks (importance:3)

Selective Data Expansion for Model Performance (importance:3)

Enhancing Offline Model-Based RL via Active Model Selection: A Bayesian Optimization Perspective (importance:3)

Solving Inverse Problems via Diffusion-Based Priors: An Approximation-Free Ensemble Sampling Approach (importance:3)

Optimal hypersurface decision trees (importance:3)

OT-MeanFlow3D: Bridging Optimal Transport and Meanflow for Efficient 3D Point Cloud Generation (importance:3)

Time-Correlated Video Bridge Matching (importance:3)

Adaptive Data Selection for Text Classification: Hierarchical Value with a Local Utility Budget (importance:3)

From Perception to Punchline: Empowering VLM with the Art of In-the-wild Meme (importance:3)

ELSA: Efficient LLM-Centric Split Aggregation for Privacy-Aware Hierarchical Federated Learning over the Network Edge (importance:3)

Maximum-distance nonnegative matrix factorization for unmixing highly mixed grain-size distribution data (importance:3)

Graph Partitioning with Demands: Generalized Conductance and its Applications (importance:3)

Mutual Information Constrained Chernoff Bottleneck (importance:3)

Universal Distributed Feature Extraction Complementary to Side Information via H-Score (importance:3)

SNACK: A Sequential Notation Framework for Probabilistic Graph Generation (importance:3)

Attention over noisy examples: generalization and memorization in the exponential Hopfield network (importance:3)

Task-aware Kernel Stein certificates for R\'{e}nyi differential privacy (importance:3)

Setting up SSH keys on macOS with Touch ID (importance:3)

Jepsen: Orkes Conductor 5.3.2 (importance:3)

The New Beings (importance:3)

Stella (importance:3)

Capitol Alpha Machine (importance:3)

Rovalty (importance:3)

Elon Musk has Grok buying him Lego sets - Mashable (importance:3)

We Asked ChatGPT, Grok and Claude for 1 Stock to Buy This Week. No One Picked Tesla. - Yahoo Finance (importance:3)

Ukraine’s drone rescue of wounded soldier puts much touted idea into practice (importance:2)

Sampling of Signals on Multigraphs: From Removable and Uniqueness sets to Spectral Proxies (importance:2)

R2-GP: Variance-Calibrated Priors for Additive Gaussian Processes (importance:2)

Regime-Aware Interpretable Ranking for Sector Rotation (importance:2)

Residual-Based Adaptive Refinement in Hybrid Variational-Collocation PINNs for Singular Elliptic PDEs with Mixed Dirichlet-Neumann Boundary Conditions (importance:2)

A Probabilistic, Rotation-Preserving Kernel Operator Surrogate for Continuum-Robot Design Spaces (importance:2)

Risk-averse learning with delayed feedback (importance:2)

Universal Learning of Nonlinear Dynamics (importance:2)

EEGDM: Label-Efficient EEG Representation Learning with Generative Diffusion Model (importance:2)

Sleep Stage Classification across Heterogeneous EEG/EOG Channels with Multi-Channel Differential Transformer (importance:2)

SPRINT: Stochastic Performative Prediction With Variance Reduction (importance:2)

Interpretable time series analysis with Gumbel dynamics (importance:2)

Empirical Bayesian Multi-Bandit Learning (importance:2)

CLAPS: Aleatoric-Epistemic Scaling via Last-Layer Laplace for Conformal Regression (importance:2)

Causal inference and model explainability analysis of a real world sales dataset (importance:2)

Variance Reduction for Stochastic Switching in Functionally Constrained Federated Optimization (importance:2)

When Does Cheap Physics Help? Gated Multi-Fidelity Physics-Informed Networks with Provable Sample Savings and Conformal Guarantees (importance:2)

Finite-Sample Wasserstein Error Bounds and Concentration Inequalities for Nonlinear Stochastic Approximation (importance:2)

$\kappa$-Explorer: Minimax Active Model Estimation in MDPs (importance:2)

PTOPOFL: Topology-Guided Personalised Federated Learning with Persistent-Homology Client Descriptors (importance:2)

RE-SAC: Disentangling aleatoric and epistemic risks in bus fleet control: A stable and robust ensemble DRL approach (importance:2)

Making a VGA Driver in SystemVerilog (importance:2)

snyvi (importance:2)

Elon Musk Grok AI Predicts an Insane Finish to 2027 for XRP - 99Bitcoins (importance:2)

MITRA-lexicon: Lexica of attested translations from Sanskrit into English, Tibetan and Chinese (importance:1)

Human Responses to Algorithmic Correction: Evidence from Centre Court (importance:1)

Quotient Geometry and Persistence-Stable Metrics for Swarm Configurations (importance:1)

DeepSeek-V4.1-Flash-1017 の「Opus 超え」を一次情報で読む ― 本当の料金、比較相手、Claude への影響 (いいね相当スコア: 0) (importance:0)

XAI Floating Rate & Alternative Income Trust Announces Preliminary Results of Oversubscribed Tender Offer - Quiver Quantitative (importance:0)

My Friend Bob Cringely (importance:0)

How to continue a GitHub Copilot app session in VS Code (importance:0)

The Thin Orchestrator: How We Made Our AI Agents More Reliable by Giving the Router Less to Do (importance:0)

Article: From Microservices to Agents: the Next Evolution of Distributed Systems (importance:0)

What are you doing this week? (importance:0)

How upload_tracker.py Keeps a Re-Run From Double-Posting Your Stock Batch (いいね相当スコア: 0) (importance:0)

The Simple Ledger That Stops Double-Posting on Adobe, Vecteezy and Dreamstime (いいね相当スコア: 0) (importance:0)

n8n Establishes Boston U.S. Headquarters and Builds Ties With Massachusetts AI Community (いいね相当スコア: 0) (importance:0)

Microsoft's October 2026 Updates Fix 1,400+ Flaws at Two Trust Boundaries (いいね相当スコア: 0) (importance:0)

Fixing Wide Tables on Mobile: Adding Horizontal Scroll without Layout Shift (いいね相当スコア: 0) (importance:0)

From Natural Language to Type-Safe SQL in Java with Querier AI (いいね相当スコア: 0) (importance:0)

Building Reliable Mixed Robot Fleets: Task Scheduling, Failure Recovery, and Distributed Intelligence (いいね相当スコア: 0) (importance:0)

Next.js 16.4 Is a Migration Release, Not a Version Bump (いいね相当スコア: 0) (importance:0)

Claude’s Scorpion-Toxin Binder Study Marks a Measured Step Forward for Protein Design (いいね相当スコア: 0) (importance:0)

How to Use Fish Audio for Beginners (2026 Guide) (いいね相当スコア: 0) (importance:0)

Free Fire Redeem Codes October: How to Claim Rewards in 2026 (いいね相当スコア: 0) (importance:0)

¿La AI solucionara todo? (いいね相当スコア: 0) (importance:0)

Rolling out Codex? Start with the workflow, not the feature list (いいね相当スコア: 0) (importance:0)

Introducing Iceberg DB: India's Serverless Vector Database & AI Agent Memory Engine (<20ms Latency) (いいね相当スコア: 0) (importance:0)

Turn Your LLM Red-Team Findings Into Code: An Open-Source Severity Scorer for Banking and a pytest CI Gate (いいね相当スコア: 0) (importance:0)

Keyword search ties a purpose-built memory layer on 200 LongMemEval questions (いいね相当スコア: 0) (importance:0)

What it actually costs to train an LLM: GPUs, days and dollars, worked out step by step (いいね相当スコア: 0) (importance:0)

Why Meta trained a small 8B model on 15 trillion tokens, and why that's cheaper than it sounds (いいね相当スコア: 0) (importance:0)

A 100K-token prompt is a denial-of-service attack you paid for (いいね相当スコア: 0) (importance:0)

Strata runs a 125B model on a 12 GB gaming GPU by leaning on system RAM, and the README spells out the cost (いいね相当スコア: 0) (importance:0)

Re-reading Nov 2024 GenAI predictions as a builder playbook (いいね相当スコア: 0) (importance:0)

Keep Claude Code, Cursor and the rest of your stack. Put Kimchi's proxy in front. (いいね相当スコア: 0) (importance:0)

he’s worried about HIS reputation 😭 (importance:0)

Weekly Self Promotion Thread (importance:0)

Harness² (importance:0)

Figma Specs (importance:0)

AIエージェントやローカルLLMを夜通し回すとき、Macのスリープで止めない方法 (いいね相当スコア: 0) (importance:0)

LINEに送った写真と音声メモを、自宅のPCが文字にしてくれるようにした (いいね相当スコア: 0) (importance:0)

日本語の指示だけでBlenderに3Dシーンを作らせてみた(mcp-for-blender) (いいね相当スコア: 0) (importance:0)

AIが回答したなら、企業の約束ではなくなるのか――Amazonのサポート対応から考える (いいね相当スコア: 0) (importance:0)

AIに仕事を任せるのではなく、管理そのものをゼロにする設計 (いいね相当スコア: 0) (importance:0)

Windows版Bloomフィルタを自作 (いいね相当スコア: 1) (importance:0)

ローカル LLM のエージェントを箱に入れても、名前の問い合わせは外へ出た。DNS の記録で気づいて止める (いいね相当スコア: 1) (importance:0)

MaixStationが面白い。NPUはいつの間にこんなに進化していたのか？ (いいね相当スコア: 0) (importance:0)

Anthropic、評価団体METRに社員並みの社内アクセスを付与 — 拒否権なしで結果を公表 (いいね相当スコア: 0) (importance:0)

コードを書かない開発者は、仕様の「抜け」をどう塞ぐか ― Cursor×Claude/Gemini×Replitの仕様駆動開発 (いいね相当スコア: 0) (importance:0)

Agent Memoryをデバッグする：KITEの構造化検索と出典の設計 (いいね相当スコア: 0) (importance:0)

AIと喧嘩して仲直りしたかったのに、心が広すぎて喧嘩にならなかった話 (いいね相当スコア: 0) (importance:0)

PythonからOpenAI APIを1回だけ呼び出す — 最小コードとGemini・Claudeとの比較 (いいね相当スコア: 0) (importance:0)

Cortex AI Function EvaluationでAI関数を評価する (いいね相当スコア: 0) (importance:0)

なぜ私のハニーポットに「攻撃AI」は来ないのか？報道された『中国製LLM×自律型AI（ARTEX）』の正体と観測の死角 (いいね相当スコア: 1) (importance:0)

承認待ちを片づけたら、比べる記録まで消えていた🧹 表示用の重複排除がモデル比較のログを毎回消していた話 (いいね相当スコア: 0) (importance:0)

AIもティルトするのか？ (いいね相当スコア: 0) (importance:0)

Diffusion Gemmaに触れる：Diffusion機構を備えたもう一つのGemma (いいね相当スコア: 0) (importance:0)

Sarashina3は「ら抜き言葉」と「なるはや」「役不足」にどう反応するか ― 文化庁『国語に関する世論調査』での検証 (いいね相当スコア: 0) (importance:0)

RecSys 2026 参加レポート：Generative Recommendation の進展と潜在的な興味の発見 (いいね相当スコア: 3) (importance:0)

AIに「仕事の進め方」を教える──Qwen3.8-27B-piの学習設計 (いいね相当スコア: 0) (importance:0)

LLMは「大きくする」だけじゃない──同じ層を回すLooped Transformer入門 (いいね相当スコア: 2) (importance:0)

ミニマックス法とは？ゲーム木探索の基本 (いいね相当スコア: 0) (importance:0)

RecSys Challenge 2026 参加レポート：コンペ参加から論文発表まで (いいね相当スコア: 4) (importance:0)

M&A 2,429件のデータから、失敗する買収を見抜くAIを作った (いいね相当スコア: 1) (importance:0)

モデルを変えずに、AIエージェントが「仕事の仕組み」を育てる──Google ResearchのHarness² (いいね相当スコア: 1) (importance:0)

QUBO++でFMQA（量子化学編）：引き伸ばしたH₂のスピン対称性の破れを60回の評価で見つける (いいね相当スコア: 0) (importance:0)

LLMスクレイパーの作り方：取得・軽量化・抽出・検証の4層設計 (いいね相当スコア: 0) (importance:0)

Python で生成 AI の引用を検証し、回答の受理条件を判定する (いいね相当スコア: 0) (importance:0)

Gemini 2.5 Flash-Liteで社内文書を整理する前に確認したいこと (いいね相当スコア: 0) (importance:0)

AIに「タスク完了」を宣言させる——証拠必須の自動クローズと、事後に覆す監査の設計 (いいね相当スコア: 0) (importance:0)

LoRAで小規模データセットから日本語LLMを追加学習する (いいね相当スコア: 0) (importance:0)

POIから作る地域特徴量＆Embedding入門：街を数値化して分かることと限界 (いいね相当スコア: 0) (importance:0)

攻撃成功率（ASR）の数字の読み方：採点基準・試行回数・データセットで同じ評価結果が 15% にも 100% にもなる (いいね相当スコア: 0) (importance:0)

【10月13日β公開】ロボットの脳はクラウドに置けるか？Physical Intelligenceの論文値で試算したら、日米の往復を足して約208msで、学習時に想定した上限200msの境目だった。推論の置き場所3択と、今すぐ試せる代替5選 (いいね相当スコア: 0) (importance:0)

AIがよく使う「効く」について (いいね相当スコア: 取得失敗) (importance:0)

【AI哲学】インターフェースの能動的システム論と実在接地メカニズムに関する総合検証 (いいね相当スコア: 取得失敗) (importance:0)

【エージェント費用7割減】200兆トークンを処理した謎の覆面AI『Space Bunny』の正体！新世代『MiniMax M3.1 Flash』の仕組みが面白すぎる (いいね相当スコア: 取得失敗) (importance:0)

【後で読む】DeepSeekの推論エンジン概説。こんなのオープンソースにするフロンティアラボはないぞ！【は馬鹿野郎】 (いいね相当スコア: 取得失敗) (importance:0)

「うちの小さなアプリなんて」と思っていたら、やられます (いいね相当スコア: 取得失敗) (importance:0)

#248：AIと数学（安野貴博の自由研究）、私の言語理論研究とLLM。。。エコー(ChatGPT)との対話 (いいね相当スコア: 取得失敗) (importance:0)

社名・商品名はトークン化で認識が変わる？表記ゆれ・造語がAI理解に与える影響 (いいね相当スコア: 取得失敗) (importance:0)

LLMは企業名をどう認識するのか？Tokenizerから始まるブランド理解の仕組み (いいね相当スコア: 取得失敗) (importance:0)

家電・スマホの”計算力”が今後狙われる (いいね相当スコア: 取得失敗) (importance:0)

10年前のAIの失敗を、いま清々しく振り返る (いいね相当スコア: 取得失敗) (importance:0)

【BTE】AIが迷わないのは、賢いから？それとも、お手上げだから？ (いいね相当スコア: 取得失敗) (importance:0)

楽天の日本語特化AI「Rakuten AI 3.0」とは｜約7,000億パラメータの国産オープンモデルの特徴と、DeepSeek V3ベース指摘の論点 (いいね相当スコア: 取得失敗) (importance:0)

LLMの遅延・コストを解決する判定特化型モデルの設計思想を学んだ記録 (いいね相当スコア: 取得失敗) (importance:0)

【AIの推論コスト】賢さは安くなるのに、最先端へ行くのは高くなる？「知能コストの裁定」を考えた (いいね相当スコア: 取得失敗) (importance:0)

AIエンジニアになるための6か月ロードマップ (いいね相当スコア: 取得失敗) (importance:0)

granite-4.2-3b-heretic 総合ベンチマークレポート（全52問） (いいね相当スコア: 取得失敗) (importance:0)

granite-4.2-3b-heretic Phase 2 日本語性能レポート (いいね相当スコア: 取得失敗) (importance:0)

AIエージェントの評価手法に関する最新動向（簡易版） (いいね相当スコア: 取得失敗) (importance:0)

無料AI APIの無料枠を比較｜Groq・Cloudflare・OpenRouterはどこまで使える？ (いいね相当スコア: 取得失敗) (importance:0)

granite-4.2-3b-heretic Phase 1 基礎性能レポート (いいね相当スコア: 取得失敗) (importance:0)

【生成AIニュース+】『FLUX 3 Max』『MiniMax M3.1 Flash』『Grok Bot』『DeepInfer』『PURE』『Strata』『CozyClay』『Ming-3.0-tts』『OpenSpline』『ComfyUI-Ming-3-tts』『MOSS-Image GGUFs』『MOSS-Image Color/Depth Control対応』『ComfyUI-Spectrum-MOSS』『MOSS Serenity』『MOSS-Image SFT DeBand』他 (いいね相当スコア: 取得失敗) (importance:0)

480GB・350億ドル・3nm。AI半導体の世界で、何か大きな変化が始まっている。 (いいね相当スコア: 取得失敗) (importance:0)

🔊音声あり（日＆英）：LLMは近所のお店を経営できるのか？最新論文「KiranaBench」をゆるく解説【AIトレンド】 (いいね相当スコア: 取得失敗) (importance:0)

障害レビューの日、「分からない」で終わらせない記録の作り方 (いいね相当スコア: 取得失敗) (importance:0)

Grok Voice 3.0を調べてみた——「世界でいちばん賢い音声AI」の数字を、AIエンジニアの目で読み解く (いいね相当スコア: 取得失敗) (importance:0)

AWS LambdaでNext.jsやHonoなどNode.jsフレームワークとアプリをそのままデプロイ、実行できる「AWS Lambda Web Functions」発表 (importance:0)

AWS Transform、Azure Functions用のコードをAWS Lambda用に変換など、Microsoft AzureからAWSへの移行に対応 (importance:0)