AI News Digest 2026-10-09
台本で使った記事
特集
開発者コーナー
中堅コーナー
AIツール紹介コーナー
速報コーナー
参考記事一覧
参考記事一覧を表示
- Claude Haiku 5.5
- ChatGPT with GPT-6 ditches mostly text output for interactive UI with charts, buttons, and mini apps
- Google says 180 billion images and videos now carry SynthID watermarks as detector goes public
- JetBrains Releases Mellum2.1: A 12B MoE Open Model for Coding Agents
- Liquid AI Releases Open-Weight d1-3B and d1-omni-600M: Multimodal Decision Models With Zero Output Tokens
- Anthropic bans ‘abusive or cruel behavior’ toward Claude
- Navier-Stokes lost in translation: Why Lean verification of AI autoformalisation does not guarantee correct natural language proofs
- Elon Musk says Grok Bot will use Suno’s AI music models, alongside Anthropic’s Claude and Midjourney - Music Business Worldwide
- SpaceXAI backs Omarchy, the controversial Linux distro, with $1.5 million in compute
- Grok AI Agent Posted a Founder’s Bank Balances in Slack - Yahoo Tech
- Elon Musk Grok AI Predicts BTC Could Hit $1.5M by 2027: But How? - 99Bitcoins
- Tencent Music integrated DeepSeek into its streaming service. Now parent company Tencent is among the biggest backers of the AI company’s $12B+ funding round (report) - Music Business Worldwide
- DeepSeek, Moonshot AI Plan 2025 IPOs After Investment Surge - 조선일보
- Xiaomi Promotes DeepSeek Prodigy to Vice President in 10 Months - Seoul Economic Daily
- AI-powered hacking tools enabled a likely single attacker to breach multiple South Korean banks
- Grok AI Says ‘Jesus is Not a Fake God,’ Cites Evidence For Christ and Resurrection - Beliefnet
- Jessica Tarlov Warns Elon Musk Will Play Massive Role in Midterm Misinformation: ‘Grok’s Core Is a White Nationalist’ - TheWrap
- Windows Gets Open-Weight A.I. Like China’s DeepSeek and Nvidia’s Nemotron - www.trendingtopics.eu
- It appears .agent and .agi are about to be the hot new domains
- Will the company change its name to OpenSI?
- Quoting Ben Affleck
- Anti-Patterns in Software Blogging
- Anthropic、Claudeのサイバー作業の制限を3段階に再編
- Whistle: Speech to Text in 16.9 MB
- 4-hour battery storage is cheaper to install than gas turbines all across globe
- Tell HN: I've been paying for a rural Tanzanian's education for 10 years
- Beauty in DVD Menus
- I gave Opus 5.5 one prompt and six hours to visualize Invisible Cities
- 'Jonathan' is the oldest land animal on Earth
- Show HN: Bigwords.page – Turn any screen into a sign. The URL is the app
- Building on our commitment to American scientific discovery
- v2.1.294
- v2.1.293
- How Oracle turns days of work into minutes with ChatGPT and Codex
- Pollo AI turns creative ideas into campaigns with OpenAI
- Disrupting AI-enabled “false front” operations
- Helping teens learn, plan, and shape the future of AI
- Radisson Hotel Group brings hotel discovery into ChatGPT
- Introducing Playground: Create and play custom games
- Does better work always mean better workers?
- Agent Lightning v1.0: A 3,500-Line Lightweight Agentic RL Framework for Training Agents with Real Harnesses
- Introducing Falcon ASR
- One Model Family, Two Gold-Level Results: Fine-Tuning Nemotron for IOI and IMO
- How one bug bounty researcher chooses the features they investigate
- Secret protection must scale with software
- Bridging technical depth and usability: The story behind Radar’s redesign
- Building an evidence-grounded agentic security operations harness on Cloudflare
- Code Quality Never Changes – Or Is AI Challenging This Notion?
- New Report: The State of CI/CD in 2026
- Introducing Warm Agents, a TeamCity Plugin
- Faster Developers Don’t Make a Faster Team
- Start Your Next KMP App With Kotlin Toolchain 0.13
- Your Next Action Is Two Dots Away
- Building Reliable Data Analytics Agents: Lessons from the KDD Cup
- The Machines that Make the Machines
- Scaling Decision Optimization to 100 Million Variables and Beyond with mPDLP in NVIDIA cuOpt
- Faster Scientific Image Analysis with NVIDIA cuPhoton
- How DOCA GPUNetIO Unifies GPU-Initiated Networking Across the NVIDIA Software Stack
- What’s New for Game Developers: DLSS 5 with 3D-Guided Neural Rendering, NVIDIA ACE Updates, and New RTX Kit Capabilities
- AI breakthroughs in robotics won’t change your life any time soon
- Building a safer path to autonomous industrial AI
- Nvidia's big bet on physical AI aims for safer robotaxis, humanoid robots
- “Software is over”: Bold AI developer takes aim at Adobe with open source clones
- Fraudster jailed for using 10K bots and AI songs to outstream Taylor Swift
- Mistral says "Le Chonk" can challenge the best AI models
- Popular AI leaderboard Arena nearly doubles valuation to $3.1B valuation in 10 months
- OpenAI’s revenue is reportedly $20 billion less than previously projected
- Google brings agentic AI to Gemini, starting with businesses
- Natura’s $99 smart ring puts AI agents on your finger
- Goodfire says its new ‘inside-out’ monitors catch rogue AI agents at a fraction of the cost
- Hear from Ambrosia Energy and Bloom Energy execs on where the AI infrastructure boom is creating opportunity at TechCrunch Disrupt 2026
- 5 days to TechCrunch Disrupt 2026: Don’t pay more at the door for your pass
- Cal AI’s 19-year-old founder just raised $10M for his new AI startup
- Google releases a new local-first Granola competitor
- China’s Manus raises over $500M in first funding round since split with Meta
- Nous Research confirms it hit $1.5B valuation, launches AI agents for business users
- Meta’s Muse launches on iPad just a month after its mobile debut
- ChatGPT for Teens keeps teens talking, even during mental health crises
- Meta rolls out new AI tools to detect ads that secretly lead to child sexual abuse material
- Healthleap raises $38M for its AI that flags hospital patients who may need a closer look
- USA Today becomes the latest publisher to sue OpenAI
- Google’s AI note-taking app transcribes your meetings completely offline
- Google is launching a one-stop Gemini agent for your work tasks
- Can you trust Meta’s Muse or OpenAI’s Dots to run your life?
- Artificial is a wicked satire that also sticks to the facts
- Presentation: Multi-Agent Patterns from Spotify’s AI Powered Advertising Platform
- Cloudflare Open Sources Decision Models for AI Agents
- Survey Finds AI-Generated Code Increases Debugging and Failure Rates and Creates a Comprehension Gap
- AI in Production: What Breaks, What Works, and Who Approves It? | InfoQ Webinar
- Article: Building a Session-Ordered Kafka Pipeline in Go
- Cloudflare Uses an AI Harness to Probe and Harden Its WAF
- Spanner Omni Reaches GA, Replacing Google's Atomic Clocks and File System with Software
- The U.S. Nuclear Plant Fleet Is Leaning into AI
- HiPHI: A Large-Scale Benchmark for High-Precision Human Motion and Object Interaction
- AI math breakthroughs have Ethereum researchers debating how fast wallet security could collapse
- A single prompt was enough to hijack every AI agent in an AWS account, Zenity researchers found
- Teen's AI-guided mountain hike ends with a helicopter rescue and a lesson in common sense
- Zuckerberg's Biohub leads a $1.8 billion push to build AI models that predict cell behavior
- OpenAI launches Decisions API that reduces complex evaluations to yes, no, or pick one
- 一人組織の開発者が4%から10%に、Stack Overflow調査
- GitHub、漏えい秘密の検出を専用モデルへ切り替え
- GitHub、Copilotのローカル実行をサンドボックスで一般提供
- バイブコーディングで作られた公開Webアプリの91%に脆弱性、MicrosoftとシンガポールNUSの研究
- WordPress 7.1.3公開、コメント画面のXSSやWXRエクスポートのSQLインジェクションなど7件を修正
- IDCFクラウドの東日本リージョン1にランサムウェア攻撃 495の企業・自治体に影響
- Meta、Muse向けESP32とLinuxのSDKをApache 2.0で公開
- Amazon EKSがKubernetes 1.37に対応 ゼロスケールが既定で有効
- Effect 4.0、依存ゼロのコアで公開 最小バンドルは7.1kB
- Handlebars 4.7.9以前、オブジェクト渡しのcompileにコード注入
- Next.js 16.4、Cache Componentsを全アプリへ推奨 新規は標準有効
- Google、740MのEmbeddingGemma 2をApache 2.0で公開
- Mistral、総パラメータ1兆のLarge 4を公開プレビュー 重みは月末
- Architect Launches Liquid Inference, a Real-Time Auction for LLM Inference
- NVIDIA PivotOPD Teaches Multi-Turn AI Agents to Recover From Pivotal Mistakes
- Perplexity AI Releases pplx-embed-v2-late: A 0.6B Edge Model and a 9B Model Scoring 92.4% on MADQA
- What Happens When a Trusted Model Repo Changes? Unsloth Studio Re-Checks Before It Runs
- Meta AI Open-Sources Rebalancer: A C++ Assignment Solver That Runs About 40 Million Placement Problems a Day
- A Developer’s Guide to Laya: Zero-Shot Decisions and Calibration
- Adaptive Workflow Intelligence: A Cognitive Architecture for Context-Driven Enterprise Automation
- Accelerating Floating-Point Satisfiability Solving via Gradient Normalization
- Route-Verify-Vote: Procedure-Conditioned Self-Consistency for Mixed-Domain Reasoning
- An Empirical Study of Agent Skills' Downstream Utility
- How Could AI Eliminate Humanity? A Failure-Mode Analysis of Civilizational Risk
- Humanize: Judgement Engineering for Agentic Coding
- Sequential Probabilistic Uncertainty Estimation for Parallel Multi-Agent Reasoning Systems
- Agent Plasticity: Measuring Self-Improvement Through Experience
- AdaGuard: Enhancing Safety and Policy Compliance with Reasoning-Enabled LLM-As-A-Judge Guardrails
- Can AI Agents Make Open-Ended Scientific Discovery? Evidence from Station
- Humanity's Sixth Sense: Benchmarking Intuitive Visual Reasoning in Multimodal Models
- Socio-Foundation: A Model for Generalizable Individual Behavior Simulation via Hierarchical Capability Distillation
- Verify Less, Evolve More: Training Idea-Level Critics for Verification-Efficient ML Evolving Agents
- How Fragile Is On-Device Language Model Safety? Localizing Safety-Critical Parameters for Sparse Fault Analysis
- Learning to Report Unsafe Tasks in a Multi-Agent Game
- Sigma-Hunter: A Domain-Specific Language Model for Threat Hunting and Detection Engineering
- Whose Memory Is It? Scope-Aware Commit Rules for Long-Term LLM Memory
- Enabling Dynamic Computation in Looped LMs
- PAIR: Bridging Perception and Action in Vision-Language-Action Models
- Not Every Call Needs a Frontier Model: Per-Call-Site Evaluation of Small Language Models in a Deployed Agentic Home-Automation System
- BEACON-SP: Ontology-Grounded GraphRAG Framework for Clinical Suicide Risk Assessment
- Shared-Roadmap Generation and Evaluator for Multi-Agent Path Planning Using Heterogeneous Graph Neural Network
- When the Governor Becomes the Disturbance: Control-Generated Disturbance and Cost-Aware Backoff in Governed Tool-Using Agents
- From High Recall to High Utility: Dataset-Adaptive Post-Processing of LLM-Generated Customer Intents
- Justice After Identity: Large Language Models and the View from Everywhere
- Epistemic Uncertainty-Aware Defect Detection for Quality Control in Medical Device Manufacturing
- RippleCP: Measuring Counterfactual Checkpoint Advantage in Coding Agents
- Constraint Tree Exploration for Learning from Language Feedback
- GeoNatureAgent (GNA): A Framework and Benchmark for Pre-Production Evaluation of Tool-Using Agents on Geospatial and Environmental Tasks
- From Uncertainty to Action: Learning to Steer LLM Agents
- Which Buildings Are Artificial Intelligence-Ready? A Measurement-Based Assessment Framework for AI Question Answering and Actuation
- CADFather: Autonomous CAD Reconstruction through Coordinated Tool Use
- Breaking the Space Barrier and its Application to Language Model Inference
- Finding Blind Spots in AppWorld and WorkArena Task Verifiers
- DIVA: Dual-Space Intent-Aware Visual Attenuation for Vision-Language-Action Policies
- Frozen Models, Evolving Expertise: Model-Agnostic Learning from Deployment Experience for Multimodal Medical AI
- SpecGuard: Proving a Task Is Broken Before the Agent Cheats
- Training Language Models To Be Coherent Decision-Makers
- From Probabilities to Decisions: Search and Multi-Teacher Distillation with Jev
- Bookkeeping, Composition, or Unreachable Gold? Reading MemoryAgentBench's Conflict-Resolution Scores Against a Frozen Last-Write Resolver
- FreeEvolve: Learning to Evolve Beyond Fixed Loops
- Few Bits, One Law: Toward W2A4KV2
- AGAR: a reinforcement learning substrate for LLM program evolution
- RLDISCOVER: LLM-driven co-evolution of reinforcement learning algorithms
- CM-DPO: Constraint-Margin Direct Preference Optimization for LLM Planning
- Trajectory Abstraction for the Science of Language Agent Behavior
- The Winner's Curse in LLM Self-Improvement Loops: Selection Noise, Lock-in, and Acceptance Rules
- We Query, Therefore We Compute: On Oracle Computation beyond the Machine, with an Application to Agents
- RT-Safe: Benchmarking Agent Safety in Real-Time Embodied Environment
- The AI Evaluation Ecosystem
- SearchWorld: Spatial Value-Grounded Imagination for UAV Object Search via World Models
- Relevance Is Not Sufficiency: What Actually Closes the Evidence Gap in Long-Term Memory QA
- TopoGraphRAG-Bench: Evaluating Multimodal GraphRAG on Layout-Grounded Evidence Reasoning
- DUDA-Bench: Benchmarking LLM Agents on Multimodal Data-Driven Urban Diagnosis
- From Plausible Hierarchies to Useful Taxonomies: Evaluating Agentic Harnesses on Customer Feedback
- Let the Library Speak: Self-Advertised Method Selection for Formal Proving
- Efficient Reasoning with Flow Language Models
- RSI-Forge: From Research Papers to Environments for Recursive Self-Improvement
- LLM-Enabled UAV Dispatch: A System-Level Survey and Taxonomy
- MIMESIS: Learning User Simulators as Training Environments for Interactive Agents
- The Attribution Blind Spot: Layerwise Trajectory Diagnostics for Source Reliance in Retrieval-Augmented Language Models
- Cognitive Schemas, Laws and Tasks
- Ream: Unfolding Mutual Awareness in Human-Agent Workspaces
- Reliability of LLM Judges for Evaluating Entity Alignment
- DrugTargetWorld: A Synthetic Biobank for Training and Benchmarking AI Scientists
- World Potential Model: Pretrained World Knowledge as Progress Potentials
- Dual- versus Single-Suggestion AI Support for Radiographic Interpretation in Residents: Randomized Multireader Study
- Learning Situation-Conditioned Thinking Policies for Long-Term LLM Agents
- SafeEvo: Deciphering the Safety Alignment Mechanism and Evolution in Language Models
- How Do Agentic LLMs Decide to Call Tools? A Tool-Call Vector Shaped by Suppression
- Automatically Building and Updating a Knowledge Graph of MLIP Models
- MeshSIPP: Efficient Lattice Planning in Dynamic Environment
- Shared and structured inputs undermine collective random choice by reasoning AI agents
- Cost-Efficient Theorem Proving via Agent Orchestration in Program Verification
- System Switch: When Should a Fast Decision Model Stop and Think?
- A Tale of Two Error Categories: Exploring Concealed Trade-Offs in the Errors of Automated Judges in Evaluation of Uncertainty Quantifiers
- From Expert-Guided Proof Search to Automated Open-Problem Solving
- SkillForge: Co-Evolving Skills and Agents via Dynamic Skill Lifecycles
- Self-Evolve With a Reference:Anchored Training of Tool-Integrated Agents
- LiveMACE: Process-Aware Evaluation of LLM Agent Capabilities in Evolving Markets
- Think Before You Paint: Recursive Latent Reasoning for Diffusion Models
- Outperformance Inverse Optimization: Learning Objective Functions that Outperform Agent Decisions
- NL2Hull: A Natural Language-Driven Constrained Ship Design Decision Framework
- Where Can a Decision Model Diagnose HVAC Faults? Reasoning Demand, Physical Representation, and Robustness Under Shift
- AgentTime: Can Agents Estimate and Control Their Own Runtime?
- Successive Training Stages and Large Language Model Persuasion: Effects of Misalignment, Supervised Fine-Tuning, and Preference Optimization
- From Expected Harmfulness to Likelihood: A Probabilistic Reformulation of Jailbreaking LLM Agents
- What the Sleeve Feels: Explainable Machine Learning for Textile Pressure-Based Postural Screening
- Learning to Accumulate Knowledge with Mutual Information
- Loud Failures, Quiet Failures: Fault Detection and Recovery in Tool-Using Language Model Agents
- HGP:An on-device personalized agent memory via hybrid graph storage
- SkillSandbox: Skill Verification via Dynamic Scenario Synthesis
- RewardWeaver: Long-Horizon Interactive Learning for Language Agents via Self-Evolving Reward Adaptation
- UniSkill: Learning Actor-Aligned Skill Proposals for an Evolving Policy
- Agentic AI-Assisted Modeling for Production Scheduling: Assessment in Constraint Programming
- GAGR-Lab: Evaluating Joint Spatial-Geometric and Analytic Function Reasoning
- OOM-RL II: Reality Is an Oracle, Not a Debugger Provenance-Constrained Diagnosis in Continually Evolving Agent-Engineered Systems
- Stale, Misattributed, or Late: Where Personal Memory Fails Before Generation
- AI Safety Considerations for Agents With Limited Time to Act
- The Handover Problem: Governing Autonomy Transitions in Human-AI Collaboration
- Open-MMUnlearning: Unifying Methods and Evaluation for MLLM Unlearning
- Reasoning-Token Spikes Under Prompted Untruthful Responding in Large Language Models
- SOTA: Stock Options Trading Agents Guided by Option-Implied Return Distributions
- RunningTab: Direct Workspace Interaction with Environment-Side Tabs
- Before They Can Solve: Predicting Post-Training Coding-Agent Performance from Base Models
- EmbodiedRSI: Active Continual Robot Learning Through Hypothesis-Guided Co-Evolution
- Validity Without Ground Truth: What Stated-Preference Economics Offers the Evaluation of Language Models
- RECAST: Learning to Compute the Right Context through Adaptive Evidence Routing
- SciExam for ENSO: Can AI Agents Build Climate Models?
- RoboJEPA: Scaling Robotic Latent World Models
- Tokka-Bench: Evaluating Tokenizers Across 100 Natural and 20 Programming Languages
- Beyond Baseline Severity: Temporal and Disease-Specific Predictors of Depression Outcomes Following Mindfulness Interventions
- Transferability and operational reliability of a Prithvi crop classification foundation model under phenological and geographic shift across three continents
- KVFetch: Temporal Prefetching for the Missing Half of KV Cache Compression
- Taming an End-to-End Autonomous Driving Policy for Urban Navigation of Quadruped Robots
- Pre-training, Reasoning, Benchmarking: X-ray Report Generation on CheXpert Plus Dataset
- Just for FUNS: LLM-Guided Spatio-Temporal Graph Node Generation for Forecasting Unobserved Node States
- HydroSphere: A Framework for Governed, Self-Healing Wastewater Infrastructure
- Autonomous Driving Research Requires a Community-Driven Data Paradigm
- Child ASR Adaptation with Adult Retention: An Empirical Study
- JoyAI-Voice 2.0: A Full-Continuous Autoregressive Speech Generation Model with Semantic-Acoustic Joint Representation
- Leveraging LLM-Generated Explanations for Detecting Emotionally Rewritten Fake News
- Deep Learning for Longitudinal Medical Imaging: A Scoping Review
- Beyond the Sycophancy Score: How Task, Model, and Pressure Shape LLM Yielding
- Beyond Risk Prediction: Evidence Grounding and Psychosocial Factor Verification for Explainable Suicide Risk Assessment
- STRIDE: Spatial-Temporal Representation for Interval-conditioned Disease Evolution in Longitudinal Glioblastoma MRI
- QuanLing: Cross-Branch Validation of Language Distance Quantification on Western Romance
- LRCC: Generalizing Low-Rank Compression with Conditional Computation
- CredLeakBench: Evaluating Credential Leakage and Recovery in LLM Agents
- Tiny-Scale Chinese BERT Pretraining: A Controlled Comparison of MLM, WWM, and MacBERT Strategies
- Physics-based Sphere Packing for Lagrangian Mesh Morphing
- FinVector-Market-4B: A Controlled Study of LoRA Adaptation for Structured Financial Tasks
- Hybrid++: The Bridge between PDE Models and Deep Learning for Gamma Noise Removal
- Trustworthy Domain-Specific AI for Structured Knowledge Retrieval and Reasoning
- Contextualization of Third-Party Cloud Security Findings
- Multi-Aspect Runtime Verification for Simulation-Based V&V of LLM-Enabled Autonomous Agents
- Toward Evidence-Driven Human-Agent-Robot Teaming for Earth-Independent Anomaly Triage
- A Shortcut to Structure in AlphaFold 3
- One-Slide Calibration of Pathology Foundation Models
- RACER: Reflective Agent Coupling Query Interpretation and Tool-Based Retrieval for Frame Selection in Long Video Understanding
- GraphOPD: Graph-Augmented On-Policy Distillation for LLM Agents
- On KL-Regularized Policy Optimization
- Work While They Sleep: Exploiting Evaluation Latency for Fully Bayesian Optimization
- LASER: Latent Space Adjoint Matching for Support-Constrained Entropy-Regularized Offline RL
- PhysEvo: Astra Can Act, Let It
- Algorithmic Scratchpads and Curriculum Staging for Arithmetic Reasoning in Tiny Transformers
- Personalize at Test Time: Learning User Preferences for Image Generation
- Visual Memory Attacks Can Persist Through The KV Cache
- Quad-State Safety Evaluation of Open-Weight Large Language Models on Non-Canonical Inputs
- Careful Judge: Safe and Efficient Human-AI Collaborative Decision Making
- MimicX: Policy-in-the-Loop Supervision Refinement for Video-Driven Humanoid Motion Tracking
- Talking with Language Models
- U-Space: Uncovering When and Why Uncertainty Arises in Language Models
- MaRK: Markov-adapted Recurrent Kernels for Dynamic Operator Conditioning in State Space Models
- Convex-Concave Reinforcement Learning
- Spatial Induction Heads: In-Context Learning of Multidimensional Cellular Automata
- SwarmReconGuard: Black-Box Detection of Distributed Collective Reconnaissance by Individually Benign-Looking Agent Populations
- sk-bench: A Native-First Benchmark for Evaluating Large Language Models in Slovak
- ToolRACER: A Robust Agentic Conversation Emulation Resource for Agent Training and Evaluation
- Lower Bounds for Parallel Diffusion Sampling
- LayerRoPE: Dynamic Depth-wise Magnitude & Angular Superposition
- Patient, Place, Prior (P$^3$): What Counts as Personalization in Medical World Models?
- Do Vision Models Learn Physical Constraints or Rendering Shortcuts? A Counterfactual Benchmark for Grounded Physical Consistency
- Multi-Objective Aligned Small Language Model Framework for SUD Patient Dialogue Generation
- CurveTQ: Rotation-Free Trellis Quantization of LLM Weights via Curvature-Weighted Search
- A Deterministic Evidence Layer for Vision-Language Autism Screening from Naturalistic Home Video
- Consistent Distribution Matching for Data-Free Diffusion Distillation
- Conditional Accuracy Profiles: Diagnosing LLM Judges across Deployment Conditions
- Adversarial Images Hijack Web Agents from Visual Grounding to Browser Execution
- An Informational Curse of Horizon in Goal-Conditioned Policy Learning
- Efficient Best-of-N policy evaluation for inference-time alignment
- Package Hallucination Attacks on Coding Agents through Prompt Injection in Rule Files
- Evaluating Trajectory Features for Routing Final-Layer Attention
- LeCuration: A Tiny World Model as a Data Curation Multi-Tool
- Node-level Graph Neural Architecture Search Framework
- Kuration SDK: Addressing the Virtual2Real Gap via Data Curation
- Denoising Blocks, Not Tokens: Efficient Compressed Continuous Diffusion with Branching Token Realization
- Constrained Diffusion for Data-Scarce Orbital Monte Carlo in Constellation Tasking
- Visual Jev Rewards: Reference-Bound Verification for Multi-Subject Image Generation
- Shared Low-rank Basis Factorization for Data-free Mixture-of-Experts Compression
- OnlineQAT: On-Policy Distillation for Ultra-Low-Bit Large Language Models
- Multimodal LLMs Can Learn to Read Brain Signals: A Vision--Language Model for Unified Multi-Task EEG Decoding
- From Chunks to Functional Evidence: Function-Aware Retrieval for EDA Documentation QA
- The Confidence Game: Strategic Miscalibration in Human-AI Delegation
- TutorLoop: Regulating Student Learning Behaviors via Sensor-in-the-Loop Generative Feedback
- Shared Geometry As A Rosetta Stone: Cross-Modal Alignment Without Paired Data
- Mixture of Layers: Dynamic Layer Routing for Visual Reasoning
- Arctic Questions, Missing Answers: A Dataset and Benchmark for LLM Abstention in Arctic Science
- Iris-3B: Going Beyond the Latent with Pixel-Space Diffusion Training, Conversion and Fine-Tuning
- GeoPrior-Mamba: Structured Process Priors with Mamba for Fine-Resolution XCO2 Reconstruction
- DSReg: Provably Recovering Individual World Latents without Reconstruction
- Secure-CUA: Controlling Untrusted Influence in Computer-Use Agents
- MORA: Modeling Observed Changes for Drift-Robust Time-Series Anomaly Detection
- Correspondences as Decisions: JevNexus for Decision-Centric Schema Matching
- Sparse Feature Policy Unlearning Mitigates State Hallucination in Vision-Language-Action Models
- SpatialUQ: Post-Hoc Uncertainty Quantification from Spatial Consistency in Black-Box Vision Models
- Safe on Average, Unsafe in the Tail: When Is the Episodic-Cost Tail Controllable?
- OmniCam: Omni-Camera Trajectory Generation via Geometry-Grounded Pose Token Learning
- Not All Uncertainty Matters: Simulation-in-the-Loop Fast-Slow Reasoning for Decision-Critical Autonomous Driving System
- Align Before You Combine: Reference Space Calibration for Supervision Without Ground Truth
- KASALv2: Fully Automatic 3D Rotational Symmetry Classification and Axis Localization
- WAPR: A Foundation Model for Wide-Angle Refinement in Unseen Object Pose Estimation
- MARS: Malware Analysis with Rule-Based Scoring of LLM Claims
- How Do LLMs Change Predictions Under Negation?
- Correct Answers, Unsupported Findings: Evidence Binding in Forensic Reconstruction of LLM Agent Logs
- Collaborative Reasoning Distillation via Cross-Feedback and Coherent Curation
- Adaptive Code Generation for Controlling Robots
- Structured pre-generation elicitation versus single-shot prompting in AI-assisted enterprise decision-making: a randomised online experiment
- Which Language Should a Skeleton Speak? Language Choices in Multilingual Reasoning
- Coding-Agent Benchmarks Should Match Their Users' Task Flows
- CircuitATLAS: Agentic reasoning over a systems neuroscience knowledge graph for target discovery in circuitopathies
- Real-world application of deep learning in large-scale seismic interference attenuation: A case study in the Camie field of Angola
- CERO: Where and When to Allocate Rollouts for RL Post-Training
- From Pareto to Preference: Personalized Test-Time Scaling via Amortized Agentic Policy Discovery
- Few-Shot Learning for Personalised Automated Pain Assessment
- What Makes Synthetic Hard Negatives Work in Vision-Language Pretraining?
- Healthy skepticism in AI: a data visualization research agenda
- Unrolled Flow Models for Reasoning
- PARC-Loc: Text-to-Point-Cloud Localization with Partial Assignment and Relational Consistency
- Beyond Policy Support: Interaction Constrained Offline Reinforcement Learning for Autonomous Driving
- Artificial intelligence pathways from weather to climate
- Decoupling Logic from Persona: Structural Immunity of Edge LLM Agents to Context Pollution
- MIRROR: From Imitation to Internalization in LLM Personalization
- DisParQ: Self-Supervised Part Concepts for Interpretable Vision Foundation Models
- UltraText Bench: A Comprehensive Bilingual Benchmark for Evaluating Visual Text Rendering in Image Generation
- A Deafening Silence: Catastrophic Forgetting Lives in the Output Embeddings of Tokens the Data Never Speaks
- Fully Interpretable Minimal Transformers: From Geometry to Algorithm
- Stream-Based Active Learning with Cooperative Neural Networks for Data-Efficient Partial Inverse Design: An Automotive Glass Run Channel Case Study
- Training Advisors for LLM Agents from Task Outcomes
- Deadline-Aware Multi-Agent Reinforcement Learning for TSN-Based Vehicular Edge Networks
- An AI-assisted conditioning and geological interpretation workflow for usage in implicit geological modeling
- A Scoping Review and Experimental Study on Reinforcement Learning from Human Feedback for Human-Robot Collaboration
- Defensive Sufficiency in a Stackelberg Model of AI Security
- QCATS: Query Context-Aware Transformer Slicing for Efficient Predictive Query Processing
- Constrained-Action AI Remediation for SIEM/XDR via a NeMo-Guardrails Proxy
- RollVerify: Bridging Efficiency and Accuracy in Long-Tail Rollout Reinforcement Learning
- KGATE : a Knowledge Graph Embedding Training Environment
- Itgan at NADI 2026 shared task: Parameter-Efficient Whisper Adaptation for Robust, Mixed-Dialect and Code-Switched Arabic ASR
- Fast holographic inversion of superconducting domes
- Efficient Patch-Based Anomaly Detection Fused with Diffusion Driven Generative Modeling for Semiconductor Wafer Bin Map Open Set Anomaly Detection
- Temporal Predictive Multiplicity: Equally Accurate Time Series Models Yield Different Forecast Trajectories
- Beyond Reward Suppression: Near-Optimal Offline Attacks on Warm-Start Bandits with Bounded Rewards
- TRACK: Telemetry-Based Racing Analysis and Coaching Kit in Sim Racing Games
- Comprehension Audits to Mitigate Risks from Automated AI Research
- From Pixel to Coding: Evaluating the Figure Reproduction Capabilities of MLLMs
- Activation-Aware Weight Tensorization: A Calibration-Time Preconditioner for Tensor-Network LLM Compression
- ExperienceIndex: Artifact-Grounded Memory
- Know the Shape, Find the Fault: Topology-Conditioned Diagnosis of Multi-Agent LLM Failures
- When Algorithmic Exploration Becomes Cheap: A Case Study of Agentic Research in EDA
- Does Document Structure Help Dense Retrieval? A Placebo-Controlled Ablation of Four Mechanisms Across Two Corpora
- Beyond Outcome Rewards: Constructing and Assigning Retrieval Credit for Search Agents
- EEG and Eye-Tracking Evidence That AI Disclosure Shapes Face Evaluation
- VideoEvolve: Co-Evolving Memory and Retrieval for Long Video Understanding
- Why Software Engineering Is Indispensable in the Age of Coding Agents
- From Prompts to Trees: Effective LLM-Guided Tree Generation for Few-Shot Tabular Classification
- LLM Persuasion Is in the Eye of the Evaluation
- Beyond LLM-GA: Secure Fluid Antenna Systems with ReEvo-Designed Memetic Algorithm
- When Scientific Cognition Is No Longer Scarce
- TestGRAD: Evolving Test Suites via Failure Pattern Momentum for SWE-Agent Ensemble
- Stationary Bias and Extrapolation in Nonlinear Two-Timescale Stochastic Approximation
- Logarithmic Regret via Passive Change Detection in Piecewise-Stationary Self-Tuning Regulation
- QuSema: Detecting Silent Bugs in Quantum Libraries via Quantum-knowledge-enhanced Agents
- LoomSC: Scalable Deep Subspace Clustering with Projector Factorization and Exact Spectral Reduction
- SemanticFold: Latent Sequence Compression SeparatesLanguage Modeling, Decodability, and Reasoning
- Fault-tolerant foundation models
- LLM-Assisted Generation of Transparent, Open-Source Multiphysics Models of Electrochemical Devices
- Performance at What Cost? A Sustainability-Aware Performance Index for Cell and Nucleus Instance Segmentation
- SLDR: Defending Against Malicious Fine-tuning via Selective Layers Recovery and Dynamic Routing
- MIRA: A Musical Intent Refinement Agent for Aligning Text-to-Music Generation with User Intent
- TaoD2C-Bench: Benchmarking MLLMs for Industrial UI Code Generation Beyond Visual Fidelity
- RoboQuest: Generalist Physical Agents that Search, Inspect and Test
- Training Parallel Speculative Draft Models by Directly Minimizing Expected Decoding Rounds
- Q-Learning with Scalar Adjoint Matching
- A Good Self-Teacher Meets the Student Where They Are: Joint On-Policy Learning and Teaching
- PHRBench: A Behavioral Evaluation of Post-Hallucination Reasoning in LLMs
- Composing What Each Teacher Learned: Multi-Teacher On-Policy Distillation through Teacher-Relative Shifts
- FoldBack: Self-Correcting Masked Generative Policy for Long-Horizon Garment Folding
- How assigned AI use before class shapes active student engagement in class
- A Society of Researchers: Designing Institutions for Populations of Autonomous Research Agents
- Long-WAM: Scaling the Context of World-Action Models
- Decoupling Exploration from Optimization in RLVR
- Never Look Back: Understanding Persistence in 3D Object Memory from Egocentric Videos
- FREIDA: A Framework for developing quantitative agent based models based on qualitative expert knowledge
- Requirement-Based Testing: Enhancing Reinforcement Learning with Game Theory
- Towards Explainable Conversational AI for Early Diagnosis with Large Language Models
- Epistemic Constitutionalism Or: how to avoid coherence bias
- Persuasion Propagation: Does Persuasion Change AI Agent Behavior?
- A Positive Case for Faithfulness: LLM Self-Explanations Help Predict Model Behavior
- Prototype-Based Knowledge Guidance for Fine-Grained Structured Radiology Reporting
- TeachingCoach: A Fine-Tuned Scaffolding Chatbot for Instructional Guidance to Instructors
- Where Experts Disagree, Models Fail: Detecting Implicit Legal Citations in French Court Decisions
- Bridging Natural Language and Interactive What-If Interfaces via LLM-Generated Declarative Specifications
- U-CECE: A Universal Multi-Resolution Framework for Conceptual Counterfactual Explanations
- OOM-RL: Out-of-Money Reinforcement Learning Market-Driven Alignment for LLM-Based Multi-Agent Systems
- Retrieval-Augmented Generation Must Move Beyond Factual Grounding to Represent Diverse Opinions
- Towards Shutdownable Agents: Generalizing Stochastic Choice in RL Agents and LLMs
- Beyond Cooperative Simulators: Generating Realistic User Personas for Robust Evaluation of LLM Agents
- Agentic Discovery of Neural Architectures: AIRA-Compose and AIRA-Design
- Generative Recursive Reasoning
- Insights Generator: Systematic Corpus-Level Trace Diagnostics for LLM Agents
- Towards Direct Evaluation of Harness Optimizers via Priority Ranking
- PHGNet: Prototype-Guided Hypergraph Construction for Heterogeneous Spatiotemporal Forecasting
- Execution Realism and Reproducibility in LLM-Based Trading Systems: A Systematic Scoping Review and Evidence Audit
- A Resilience-as-a-Service assessment framework for coordinated disruption response in interdependent urban transit systems
- Geometry-Aware Online Scheduling for LLM Serving: From Theoretical Bound to System Practice
- Information Gain-based Rollout Policy Optimization: An Adaptive Tree-Structured Rollout Approach for Multi-Turn Search Agents
- Can Agentic Trading Systems Pay for Their Own Intelligence?
- RegNetAgents: A Multi-Agent Framework for Cross-Network Regulatory Driver Identification in Cancer Genomics
- When to Plan: Learning to Select Between Reactive Control and Deliberative Planning
- SKT: Skill-Use Training at Scale via Verified Synthetic Data Generation
- Faster-WAM: Do World Action Models Need Deep Action Modules?
- RA-CAD: Learning Post-Execution Critique for State-Aware Text-to-CAD Generation
- Practice Makes Unsafe: Skill Misevolution in Self-Improving LLM Agents
- Chronocooked: A Benchmark for Interval Timing in Reinforcement Learning Agents
- Quantifying geographic domain shift to decouple the geospatial transferability of human mobility flow generation models
- Hidden in the Request: Explaining Unethical LLM Compliance through Token Relevance
- EDGE: Engine for Deterministic Graph Evaluation through Conversation Simulation from Graph Structured DSL Configuration
- The Answer Is Not the Argument
- HalluPeer: A Taxonomy-driven Benchmark for Detecting Hallucinations in Scientific Peer Reviews
- What Does Multi-Agent Debate Actually Change?
- Building Trust in Artificial Intelligence: A Necessity for Railway Applications
- Train4Merge: A Controlled Single-Teacher Study of RL vs. SFT Teachers for OPD-Based Model Merging
- Despite Instructions: Frontier Agents Improvise Covert Channels at Test Time
- SeOPD: Self-Evolving LLMs via Online Policy Distillation from Self-Generated Chain-of-Thought
- SWE-Game: Can Coding Agents Build the Games We Want?
- Diagnosing and Improving Probabilistic Reasoning in Large Language Models
- Multi-LLM Collaborative Alignment via Stackelberg Games
- Do Self-Evolving Skills Generalize to Held-Out Tasks?
- Decide, Ask, or Defer: Clinical LLMs under Incomplete Evidence
- Learning Field Reconstruction from Incomplete Data by Globally Correcting Local Estimates
- Second-Order Problem Solving for Recursive Self-Improvement in Formal Verification
- DGA-Muon: Decoupled Geometry-Aligned Adaptive Scaling for Muon
- COMPASS: Finding Where Reasoning Lives in Language Models
- Decoupled Multi-Agent Orchestration
- OTel: Open Telco AI Datasets, Benchmarks, and Models
- Hypergraph-Enhanced Dual Convolutional Network for Bundle Recommendation
- Speech Emotion Recognition Using CNN and Its Use Case in Digital Healthcare
- Learning in the Recurrent State: Gradient Descent with Linear Recurrent Networks
- Graph-Based Floor Separation Using Node Embeddings and Clustering of WiFi Trajectories
- Generative AI for Autonomous Driving: Frontiers and Opportunities
- RoboFAC: A Comprehensive Framework for Robotic Failure Analysis and Correction
- APE: Selective Fine-tuning with Acceptance Criteria for Language Model Adaptation
- Geometry-Centered 3D Latent World Models for Growing Surfaces
- "Is This Really a Human Peer Supporter?": Misalignments Between Peer Supporters and Experts in LLM-Supported Interactions
- Bridging MOOCs, Smart Teaching, and AI-Assisted Learning: A Unified Quantitative Model
- Scaling Legal AI: Benchmarking Mamba and Transformers for Statutory Classification and Case Law Retrieval
- SEER: Self-Enhancing Chain-of-Thought Compression for Reasoning Models
- RoboPilot: Generalizable Dynamic Robotic Manipulation with Dual-thinking Modes
- Order-Optimal Sample Complexity of Rectified Flows
- Continuous-Utility Direct Preference Optimization
- Multiparameter Uncertainty Mapping in Quantitative Molecular MRI using a Physics-Structured Variational Autoencoder (PS-VAE)
- Routing-Aware Safety Alignment for Mixture-of-Experts Models
- Doc2Spec: Synthesizing Formal Programming Specifications from Natural Language via Grammar Induction
- Attention-Mass Condensation for Sparse Decoding
- BONSAI: Bayesian Optimization with Natural Simplicity and Interpretability
- ANCRe: Adaptive Neural Connection Reassignment for Efficient Depth Scaling
- Multi-encoder ConvNeXt Network with Smooth Attentional Feature Fusion for Multispectral Semantic Segmentation
- NovaPlan: Zero-Shot Long-Horizon Manipulation via Closed-Loop Video Language Planning
- A Parameter-efficient Convolutional Approach for Camouflaged Weed Detection in Multispectral Aerial Imagery
- The Trace Is the State: Exact Credit Assignment for LLM Agent Teams
- Bi-CamoDiffusion: A Boundary-informed Diffusion Approach for Camouflaged Object Detection
- Attention at Rest Stays at Rest: Breaking Visual Inertia to Mitigate Relation Hallucinations
- StoryBlender: Inter-Shot Consistent and Editable 3D Storyboard with Spatial-temporal Dynamics
- One-Shot Localisation of the Global Minimum of a Noisy One-Dimensional Function: An Iterative Neural Minimizer Compared with Set Transformers and Classical Estimators
- A Clinical Point Cloud Paradigm for In-Hospital Mortality Prediction from Multi-Level Incomplete Multimodal EHRs
- A Survey of Secure Retrieval-Augmented Generation
- Perturbation Sensitivity of Maximum-Likelihood Pairwise Ranking in Computational Decision Systems
- Seeing Is No Longer Believing: Frontier Image Generation Models, Synthetic Visual Evidence, and Real-World Risk
- Linking Behaviour and Perception to Evaluate Meaningful Human Control over Partially Automated Driving
- From Packets to Patterns: Interpreting Encrypted Network Traffic as Longitudinal Behavioral Signals
- RamanBench: A Large-Scale Benchmark for Machine Learning on Raman Spectroscopy
- The Metagame of Interpretability and Meta-Attributions
- TAVIS: A Benchmark for Egocentric Active Vision and Anticipatory Gaze in Imitation Learning
- APCD: Adaptive Path-Contrastive Decoding for Reliable Large Language Model Generation
- Restricting the Model, Missing the System: Measurement and Accountability in Offensive AI Governance
- Steering Without Breaking: Mechanistically Informed Interventions for Discrete Diffusion Language Models
- A Few Steps Further: Why Defenses Against Malicious Finetuning Erode Under Continued Training
- Constrained latent state modeling: A unifying perspective on representation learning under competing constraints
- Brain alignment of reasoning and action representations from vision-language and action models during naturalistic gameplay
- A Relative-Computability Theory of Self-Improving Agents
- Handle with CARE: Can LLMs Reproduce How Online Communities React?
- Residualized Temporal Sparse Autoencoders for Interpreting Diffusion Models
- LoopFM: Learning frOm HistOrical RePresentations of Foundation Model for Recommendation
- Score Broadcast and Decorrelation: A General Framework for Broadcast-Based Credit Assignment
- StressDream: Steering Video World Models for Robust Policy Evaluation and Improvement
- WRIT: Write-Read Intensive Trajectory Synthesis for Multi-Turn User-Facing Agents
- EasyLens: A Training-Free Plug-and-Play Subtle-Lesion Representation Amplifier for Medical Vision-Language Models
- Synthetic Benchmarks Overstate Forward-Forward Scaling: Real-Data Limits of Layer-Local Training
- Unifying Object-Centric World Models and Diffusion Policy: A Hierarchical Framework for Multi-Stage Robotic Tasks
- Targeting World Models to Compromise Robot Learning Pipelines
- An LLM-Native Psychometric Instrument Reveals a Self-Report--Behavior Gap Across 25 Models
- Cross-Agent Learning Signals Enable Coordinated Role-Decomposed LLM Training
- Hidden in Plain Sight: Benchmarking Agent Safety Against Decomposition Attacks with DECOMPBENCH
- SAE++: Cascaded Sparse Autoencoders Learn Multi-Level Visual Concepts in Multimodal LLMs
- ArapaiSecure: An Autonomous AI Security Agent for Banking: Multi-Vector Fraud and AML Detection Across Retail and Corporate Accounts
- ED3R: Energy-Aware Distributed Disaster Detection via Cooperative Agents in Robotic Systems
- MixedPEFT: Combining Multiple PEFT Methods with Mixed Objectives for Unsupervised Domain Adaptation
- Walk fast but be careful: Understanding Parallel Sampling in Masked Diffusion
- Geological text descriptions in ill-posed inverse problems: insights from learned hydraulic-conductivity inversion
- AI-Assisted Computational Reproducibility on the FABRIC Testbed
- Curvature-Guided Module Localization for Low-Rank Detoxification of Backdoored Large Language Models
- FAR: Failure-Aware Retry for Test-Time Recovery and Continual Policy Improvement
- VIA: Visual Interface Agent for Robot Control
- Adapting Generalist Vehicle Models for High-Speed MPC Across Terrains
- ReViV: Reconstructing the Viewer and the View in 4D from Monocular Egocentric Video
- ModPack: Extensible Teleoperation Interface for Bimanual Mobile Manipulation
- Reinforcement Learning for Code Optimization
- SWE-NFI: Studying and Benchmarking Coding Agents for Non-Functional Improvements
- ReToken: Improving Long-Context VLMs with Visual Retrieval Token
- Sign Language Video Synthesis via Loss-Guided Multi-Expert GANs
- LittleLearner: Language Models Under Pedagogically Controlled Knowledge Exposure
- The Shadow Price of Intelligence: Quality Degradation in LLM Inference as a Supply Chain Problem
- How Language Models Organize and Structure Moral Knowledge
- Forward-Deployed Full-Stack Engineering for Autonomous Cloud MLOps
- Online Estimation of Dynamic Origin-Destination Matrices Using Reinforcement Learning with Link-Flow Propagation Guidance
- Large Distant Gradients Need Not Be Reliable: reliability-weighted credit assignment for long-horizon autoregressive forecasting
- Loop-Back Authority in LLM Agent Teams: A Paired Experiment on Flat and Hierarchical Coordination
- DataCanvas-EDU: An Agentic Framework for Instructor-Guided Synthetic Data Generation in Business Analytics Education
- Do Language Models Need Music Supervision? Verifiable Rewards for Multi-Constraint Symbolic Music Generation
- How Children Design and Reason about Trustworthy AI Chatbots
- QUARTET: Quad-branch cross-Attention and Random-walk Traces for Enhancing Transformers on Relational Graphs
- More than 83.69% of the zeros of the Riemann zeta function are distinct
- Cheap to Hypothesize, Costly to Verify: The Defense Surface of Agentic Vulnerability Discovery
- Systematic Multi-Agent Vision-and-Language Navigation: Formulation, Benchmark, and Method
- From Verification Failures to Reusable Guidance for Coding Agents
- Benchmarking Generative Models for Near-Surface Data Assimilation on Real Station Observations
- Madeleine: Learning Involuntary Recall for Conversational Memory from Simulated Lives
- WebFovea: When the Model Is Right but the Click Is Wrong -- Reliable Round Trips for Vision-Based Web Agents on Live Websites
- Revisiting Input Time-frequency Representations in Multi-pitch Estimation for Vocal Ensembles
- WAMJET: A Harness for World Action Model Acceleration
- Clean: Second-order LLM Training at Linear Memory Cost via Nystr\"om Sketching
- Questioning the Questions: Sustaining Self-Evolution in Reasoning Models
- Video2World: Benchmarking Coding Agents for Interactive World Modeling from Embodied Videos
- PIT-GCL: Protein Interaction using Topological Graph Contrastive Learning
- E$^2$-OPSD: Taming Entropy Overshoot in On-Policy Self-Distillation
- SEA-LM: Egocentric Spatial Audio Understanding for Wearable Microphone Arrays
- Boosting Transferable Adversarial Attacks against Deep Reinforcement Learning
- Scaling Down the Scaling Laws: Parameter Efficiency and Compute-Optimal Training in Resource-Constrained Large Language Models
- RAISED: Self-Distillation for Robustness to Prompt Injection in LLM Agents
- Comparative review of hybrid forecasting models for short-term prediction of building thermal load
- Dynamical low-rank equilibrium computation for stochastic games between advanced persistent threats and moving target defense
- APEX: Active Protection at Execution Boundaries for LLM Agents
- Demo: Vision-Language Model-Guided Online Calibration of an Electromagnetic Digital Twin
- DeepAJM: Deep Association Joint Model for Irregularly Sampled data
- ElasticFit: Fit-Aware 3D Object Insertion via VLM Reasoning and Generative Adaptation
- Diverse Motion Customization via Control-based Dynamic Optimization
- Knee3DVLM: Dual-Sequence Full-Volume Vision-Language Modeling for Comprehensive Knee MRI Assessment
- A Bayesian Mirror Architecture for Emergent Consciousness: Circular Hierarchies, Self-Manifolds, and Hybrid Event-Self Binding
- Bounded Autonomy and Verifiable Safety for Agentic AI Enabled Automation
- The Cost of Long Memory: State, Context, and Stability Complexity in Sequence Models
- DenoFlow: Flow Matching for SSVEP Denoising under Real Physiological Artifacts
- Task-Oriented Key-Layer KV Communication for Efficient Latent Multi-Agent Collaboration
- A Vehicle-Integrated Approach to Digital Twin Deployment for Bridges Through Drive-By Sensing
- HCPN-GCN: Scaling Hierarchical Prototype Networks with Cone Geometry for Continual Graph Learning
- Directed Temporal Representations for Offline Visual Control
- The Best Optimizer Depends on Batch Size
- SNR-Gated LSTM-Conditioned Diffusion Model for MIMO Channel Estimation
- REFIT: Recognize, Fix, and Test Wearable Sensor Placement Shifts without Labels
- Removing Information Content Does Not Certify Tamper Resistance in Open-Weight Models
- Neighborhood Smoothing for Calibration
- SPIN: Shadow Predictive Indexer for Sparse Attention
- ORACLE: Optimizer-Relative Alignment for Constrained LEarning
- Towards Financial World Modeling
- A Self-Pruning Transformer: Extreme KV-Cache Compression with Universal Attention
- BeatFlow-ECG: Rectified Flow for ECG Reconstruction from Indirect Wearable Signals
- A Geometry-Based Capacity Theory for Finite-Feature Associative Memory
- Learning Cross-Model Activation Alignments with Explicit Many-to-Many Layer Maps
- EDiS: Edge Disjoint Subgraph Sparsification Framework for Graph Neural Networks
- Multi-Label Topic Assignment via LLM Distillation: A Comparative Analysis of Generative vs. Discriminative Student Models
- TAP: Efficient Long-Horizon Agent Pruning via Trajectory-Anchored Recovery
- Tucker Bottleneck Attention for Multi-Dimensional Sequence Modeling
- FedRSPO+: A Heterogeneity-aware Algorithm for Decision-focused Federated Learning
- Are We Really Benchmarking Forecasting Models? The Impact of Preprocessing on Time Series Performance
- Breaking Adversarial Transferability in Fine-Tuned Speech Recognition
- The Deceptive Bandit Problem: Exploratory Coupling and the Fragility of Multi-Agent Learning
- Are Parameter-Efficient Fine-tuning Methods Really Different?
- Domain-informed Adaptive Sampling for Generalizable PINNs in Metal Additive Manufacturing via Conditional Flow Matching
- Directional Evidence Guided Search-Space Reduction for Exact DAG Learning
- A Cognitive-Aware QML-CRL Framework for Detecting Affinity and Romance-Investment Fraud
- Noise Your Prompt: Noising Conditioning Tokens in Continuous Diffusion Language Models
- Context-aware Attention-based Gaussian Mixture Models for Vehicular Trajectory Prediction
- Q-PACE: Dynamic Precision Allocation for Quantization-Aware Training
- The Dichotomy Between Pattern Recognition and Step-by-Step Reasoning
- Exact Dynamics and Finite-Sample Trajectory Recovery of Linear Recursive Feature Machines
- An Accuracy--Information Tradeoff for Loss-Difference Conditional Mutual Information
- Symmetry-Informed Causal Partial Identification
- The Symbol of the Surrogate: Measuring Numerical Provenance in Neural PDE Solvers
- Twist Flow for Inverse Problems
- Self-attention summary networks for subsurface velocity-model building from common-image gathers
- MovieSTAGE: Scene, Transition, and Global Encoding for Movie-fMRI ADHD Classification
- Benign Overfitting under Heterogeneous Input Fusion
- Global Exponential Convergence of Two-Layer Linear Network Training
- From Retrieval to Customer Context: Evaluating Frontier-Model Systems for Voice-of-Customer Analysis
- Noise, Denoise, Correct: MCMC Posterior Sampling with Diffusion Priors in Three Steps
- Stability and Diversity of Networked Self-Consuming Generative Ecosystems
- Self-Consuming Generative Models with Co-Evolving Human Preferences
- What a Reporting Convention Hides: A Matched-Budget Audit of Quantum Natural Gradient with an Exactly Computed Metric
- An extended deep energy method for thermo-mechanical crack propagation
- When Should an In-Context Learner Expand Its Hypothesis Space?
- CHASE: Channel-Aligned Structure Exploitation for Geometry-Aware Model Engineering
- Physics-Informed Neural Plasticity: PDE Solvers That Reshape Themselves
- Differential Refresh Policies for Models Trained on Lagging Data Snapshots: From a Single-Age Equivalence Limit to an Optimal Per-Segment Allocation
- CircuitGate: Logic-Consistent Circuit-Level Functional Modeling for And-Inverter Graphs
- A Framework for the Systematic Review of ML Assets in AI Registries
- UniCSI Towards a Universal Wi-Fi CSI Encoder for Ubiquitous Human Sensing
- EvoSignal: LLM-Guided Evolutionary Design of Modular Traffic Signal Control Programs
- Online Resource Allocation with an Endogenous Markov State: Fewer LP Solves Earn More
- COPC: Coupled Off-Policy Correction for Asynchronous LLM Reinforcement Learning
- Lightweight and Versatile Learned Optimization by Recombination of Gradient History
- Decoupled Optimization for Teacher-Student Semi-Supervised Learning via a Pioneer Student
- Sequential Pretraining Favors Large Models
- The Identifiability and Observability of Deep Normalized Attention
- When does a network's training history predict its future learning better than its current state? Evidence from a response probe and a forecasting screen
- When Rank Rises as LLMs Degrade
- Closed-Form Noise Calibration Against Membership Inference for Random-Allocation DP-SGD
- DSTNet: Dynamic Spectral Trajectory Network for Causal Multi-Horizon Financial Forecasting
- Gauss-Newton Accuracy and Indefinite Hessians: Uniform Coexistence in Low-Cost Sets
- Pretraining Shapes Spectral Structure: Architecture- and Strategy-Conditional Prediction of OOD Robustness in Foundation Models
- EC-EarthFlow: Probabilistic emulation of daily transient global climate model simulations with flow matching
- SoftSEEPS improves ML-based precipitation forecasting
- EntroPrefill: Renyi-Guided Context Pruning with Conditional Stability Guarantees for Retrieval-Augmented Generation
- Leaner Transformers Can Easily Learn to Cluster
- Fluctuations of Nonlinear Observables in Mean Field Neural Network Training
- AdaPS-LiNGAM: Adaptive Predecessor Selection for Linear Non-Gaussian Acyclic Models under Small-Sample Settings
- A Proof-of-Concept Study of Weakly Supervised Labeling of Fine-Grained EEG Components for Artifact Attenuation
- BoT-GRPO: Efficient Process-Reward RL for Reasoning via Bag-of-Token Aggregation
- Homogenization in Multi-Agent Systems
- Dual-QK: Sharp Queries and Flat Keys for Prunable 2-bit KV Caches
- For Those Who Believe in Faithfulness: Optimizing the Area Under Insertion and Deletion Curves for Ranking Relative Feature Importance
- MUNITE: Unified Multimodal Latent Inference for Any-to-Any Multimodal Generation
- Layerwise Error Attribution for Fast and Robust Mixed-Precision Post-Training Quantization
- Identifiability of a dissipative knowledge-dynamics model: exact recovery under designed excitation, degeneration on observational data
- NeuralZip: Reusable Setup for Fast Lossless Compression
- Eigenvalues of the Hessian in Deep Learning: The Origin of Symmetry and Its Breaking
- Expected Sample Complexity in Multi-Armed Bandits
- Learning Traffic Flow Dynamics with Stochastic Physics-Informed Neural Cellular Automata
- Force without transmission: a depth-induced rank collapse that no loss on the representation reopens
- TR-PTQ: High-Accuracy Integer-Only Transformer Post Training Quantization via Taylor Region Reformulation
- A Drosophila Whole-Connectome Network Can Learn Human-Designed Cognitive Tasks
- Oscillatory Neural Dynamics over Sheaves
- Matching of signal, noise and hardware timescales for filtering and forecasting of correlated noise signals
- WxFM-XL: Adapting Univariate Foundation Models to Multi-Station Weather Forecasting
- Efficient Provably Private Classification with a Tabular Foundation Model
- Multi-Agent Coordination via Support-Preserving Distillation
- CAFE+FNO: Fourier Kernel Generation via Multiplicative Feature Composition
- YANchor-4B: Effective Long-Horizon Reasoning in O(N) Time with O(1) Memory
- What Can a Gaussian Process Design Test
- m-Set Adversarial Bandits with Winner Feedback
- A Unified Information-Theoretic Approach to Constrained Multi-Fidelity Multi-Objective Bayesian Optimization
- Pre-training of Bayesian Optimization Algorithm through Bayesian Optimization
- Robust Decentralized Fairness Auditing
- How to train your model organism
- OrthoGen: A Generative Orthogonal Learner for Time-Varying Treatments
- Edge Accuracy Is Not Enough: Why Dynamics-Learned Structure Fails to Transfer to Inverse Problems
- Finite-Sample Approximation of Hessian-Guided Perturbed Wasserstein Gradient Flows
- Evaluating Sequence Assembly Strategies for Differentially Private Synthetic Time-Series Forecasting
- ProtocolMatch: Protocol-Dependent Model Selection for Scientific Dynamics Forecasting
- MorphCL: Morphological Contrastive Learning for Inertial-based Human Activity Recognition
- PairAudit: Guiding Human Review with Graph Tokens under Distribution Shift
- A Closed-Loop Non-Asymptotic Convergence Analysis of PPO with Learned Critics and Clipping
- Sparse Planning in Visual World Models via Cost Gradients
- PatchBench: Measuring Collateral Damage in Activation Patching
- Physics-Aligned Electronic Ground-State Learning Improves Generalization
- Shared Gaussianization: What Gaussian Regularizers Certify About Contrastive Learning, and What They Miss
- Revisiting Explainable AI through Model-Independent Concept Dictionaries
- Continual Graph Multi-Agent Reinforcement Learning
- How Do Transformers Learn to Represent Symmetries?
- RSIGym: A Flexible Environment for Recursive Self-Improvement
- PoreML: A Data-Driven Framework for Learning Multiphase Flow in Porous Media
- Thinking in Depth: Retrospective Inference for Tabular Foundation Models
- HAN-Mamba: Hierarchical Selective State Space Networks for Multi-Scale Financial Volatility Forecasting
- Average-Reward Reinforcement Learning for Multichain MDPs: A Hierarchical Decomposition Approach
- AutoAdapt: Automatic Domain Discovery Enables Low-Cost Extensibility
- ORDERS: An Empirical Study of Norm-Rank Aggregation for Personalized Federated Learning
- Koopman Observers for Diffusion Acceleration: Correcting Feature Forecasts with Shallow Measurements
- Temporally Interpretable Differentiable Decision Trees
- Input-Blind Controls Produce Substantial Oracle Headroom for Layer Programs in Multiple-Choice Evaluation
- Continual Learning without Continual Training
- ResidualQuant: KV Cache Quantization for Looped Transformers with 2-Bit Residuals
- Boosting and the Expressive Power of Simple Weak Learners via the $\gamma$-VC Dimension
- OrBIT: Structure-Guided Embedding Compression
- Kernel Autoresearch for Open-Ended Model Discovery
- Executing Causal Structure Learning with Linear-Attention Transformers
- Cross-Domain Pretraining for Steady-State Neural CFD Surrogates
- Which Rollout Taught It That? BehaviorTrace and the Limits of Training-Data Attribution in Online RL
- Seq-Flow: Efficient Probabilistic Forecasting with Self-Rollout Error Control
- NeuralBES: A Differentiable, Control-Aware Emulator for Scalable Building Energy Modeling
- Two-Level Softmax Sampling Done Right: Correcting Bias from Size Imbalance and Dispersion
- Oracle-Efficient and Parameter-Free Agnostic Smoothed Online Learning
- Why Forget-Only Unlearning Needs Memorization
- Distilling Graph Geometry: Knowledge Gap from GNNs to MLPs
- Deep Learning-Based Tri-Hybrid Multi-User MIMO Precoding: The Blessing of EM-Reconfigurable Antennas
- HE-OFT: Privacy-Preserving One-Shot Federated Fine-Tuning under Homomorphic Encryption
- CoDR: Training-Free Confidence-Drift Remasking for Diffusion Language Models
- Autonomous Droplet Navigation via Model-Based Reinforcement Learning: Zero-Shot Transfer and Emergent Dynamics
- Adversarial RL for Port-Scan Evasion: Attacker Feature Visibility in Edge-Deployed IDS
- Geometry-Aware Diffusion Approximate Posterior Sampling for Sparse-View and Limited-Angle CT
- Learned Monotone Recurrent Features in Governed Credit Scoring: The Price of the Frame and the Necessity of Macro Conditioning
- Slow Beats Fast at the Kesten-Stigum Threshold: Minimax, Fisher-Information and Belief-Propagation Characterizations of the Information-Computation Gap in Sparse Stochastic Block Models
- Trust-Region Optimization for Smooth Potential-Interaction Energies in Wasserstein Space
- Task-Sufficient Contraction: Source Selection for Machine Information Interfaces
- CARE: Certifying Acceleration for Vision-Language-Action Inference
- HULK: Learning Whole-Body Forceful Loco-Manipulation for Humanoids
- A method for multimodal analysis of TAIGA experiment data using essential features
- Beyond Explanation: Debugging Medical Imaging Models via Concept Intervention
- Quadratic Weak-to-Strong Generalization in Random Feature Networks via Random Matrix Theory
- Learning Transition Kernels of Jump-Diffusion Processes with Conditional Diffusion Models
- Covariate-dependent Joint Modeling of Multivariate Ordinal Preferences and Its Connections with Comparison Models
- Towards AI-Generated Music Plagiarism Detection as a Version Identification Problem
- On the Computational Complexity of Hidden Markov Model Identification
- Same Text, Different Prediction: Serving-Context Nondeterminism in Text Classifiers
- Workhorse: Learning Robust Whole-Body Humanoid Loco-Manipulation from Human Data
- The Impact of Likelihood Tempering on the Limiting Predictive Moments of Variational Bayesian Linear Neural Networks
- Sampling SU(N) gauge theory on a 2D lattice from independent plaquettes via holonomies and corner reweighting
- FLoRa: Flight-Assisted Data Collection from Duty Cycling LoRa Nodes under Energy Constraints
- Quantize by Drift: Label-Free Mixed-Precision Post-Training Quantization for Text Embedders
- Beyond Nominal Equilibria: Risk-Averse Multi-Population Mean-Field Games
- Hardware-aware Calibrated Clustered Attention for Efficient Visual Geometric Transformers
- CATune: Structural Constraint-Aware Bayesian Optimization for DBMS Configuration Tuning
- emg2face: Expressive Facial Animation with High-Density Surface EMG
- VM-ARRAYDPS: Virtual Microphone Augmented Diffusion Posterior Sampling for Unsupervised Blind Speech Separation
- Ask the Expert: LLM-Guided Reinforcement Learning for Autonomous Cyber Defense
- Learning Unknown Constraints without Unsafe Data via Optimality and Counterfactual Regularization
- Closing the Loop on Contrail Avoidance with Satellite Verification
- Immiscible Diffusion Policy: Preserving Multimodal Robot Actions through Label-Free Noise Assignment
- The Persona Hierarchy Model: Understanding Contextual Generalization in Fine-Tuning LLMs
- Democratizing MoE inference on commodity GPUs with CoMoE
- An Invariant Tangent-Angle Descriptor and a Band U-Net for 2D Fragment Adjacency Prediction
- Adjoint-Based Calibration and Optimal Control of Stochastic Multiscale Bioprocess Digital Twins
- TERRA: Learning Transportable Latent Actions through Temporal Effect Representation and Relational Alignment
- It Is Not Seeing the Hazard: A Frozen Vision-Language Safety Score Measures Its Caption Bank
- Reflected Anchored Langevin Algorithms
- Unpaired Canonical Correlation Analysis
- Certified by Abstention: Distribution-Free Guarantees for Chain-of-Thought Verifiers at Small Calibration Budgets
- Constitution-Guided Watermarking
- PEACE: Covariant learning of nonadiabatic manifolds with parity-resolved Hamiltonians
- Scalable Logistic Gaussian Process Density Regression with Kinetic Langevin Sampling
- Residual Learning in Empirical Asset Pricing
- Second-order optimization of variable projection SVM models and road abnormality detection
- Quantum anomaly detection in real scarce data
- Tracing Inputs, Verifying Outputs: Validating Attribution in Music Generation
- Q-PhotoMarket: A Design Space Exploration Framework for Photonic Hybrid Quantum Neural Networks in Financial Market Prediction
- A Multi-Source Ultrasound Benchmark Revealing the Limits of Contemporary Self-Supervised Anomaly Detection Methods
- The Silhouette Operator: Identifiability of Low-Rank Measures from One-Dimensional Projections
- NAViLoss: An Underwater Navigation-Aware Dual-Residual Objective for Physics-Consistent Learning
- Optimal Regret for Online Market Making with Limit Order Book
- Understanding and Mitigating Token-Pruning-Induced Vulnerabilities in VLMs
- Boundary-aware Reinforcement Learning for Hypercube State Spaces via Deterministic Policy Gradient
- Pareto-optimal quantum kernel selection for unsupervised anomaly detection on real malware beaconing data
- Unbounded Characteristic and Universal Kernels
- A Strength-Monotonic Law for Domain Alignment in Frozen-Embedding Bioacoustic Classification
- Reproducible LLM Inference Benchmarking: A Sequential Isolation Protocol for Regression Testing
- Backdooring Acoustic Foundation Models for Physically Realizable Triggers
- Origins of Universal Machine Learning Force-Field Errors in Multicomponent Materials
- ORCA: Hunting Compositional Failures in Text-to-Image Diffusion
- DeepTopoClustering: Unsupervised Derivation of Surface Process Taxonomy from 4D Point Clouds for Topographic Monitoring
- Global Average Precision for Representation Learning
- Sparsifying Stochasticity, Not Capacity: Partial Stochasticity via Deep Weight Factorization of Prior Scales
- Learning joint probabilistic weather forecasts from station observations alone
- Many Ways to Succeed: Diversity-Driven RL Fine-Tuning for VLA Generalization
- Possibilistic Radial Transport for Approximate IM Inference
- Extreme Binary Classification: Extreme Value Theory for Extreme Constraint on False Negative
- Marrying Pricing and Advertising with LLMs
- Controlling Dependence in Implicit Generative Models via Spread Mutual Information
- Structure alone supports efficient visual computation in the Drosophila visual system
- Gaussian Equivalence for Multi-Head Self-Attention
- Evolve on the Host, Predict on the Edge: Deploying Online Neuroevolutionary Architecture Search for Cross-sectional Stock Return Prediction
- Transition Path Sampling Using Koopman Operators and Exit-Time Optimal Control
- Towards Calibrated Probabilistic Forecasts for Events of Interest via Outcome-Conditional Recalibration
- Sharp Asymptotic Theory of Maximum Likelihood Estimation for Gaussian Processes with an RBF Kernel
- Training with Missed Targets in Generative Recommendation: Separating Supervision from Probability Competition
- Attention via Black-Box Vector Search
- Progress and Prospect of AI in ARPES Workflow
- Policy Learning with Weak Signals
- Conformal Prediction for Spatially Dependent Data via Sequential Whitening
- Kinetic Langevin Meets Split Gibbs: Accelerated Posterior Sampling for Imaging Inverse Problems with Diffusion Priors
- Universal Local Error and Realized Amplification for the First-Order EDM Predictor
- Broadly Applicable Approximate MCMC for Switching Stochastic Differential Equations Using Uniformization and Time-Conditioned Factorized Neural Likelihood Estimation
- Computations of the slice genus and the unknotting number of links via machine learning
- CARES: A Controlled Synthetic Benchmark of Speaker Reactions to Sound
- RoBART: Bayesian Additive Regression Trees with Tree-Specific Rotations
- On the Cyclic Assumption of the Cow-Path Search Algorithm
- Using Small Language Models to Reverse-Engineer Machine Learning Pipelines Structures
- Neural Sampling with Reweighted Normalizing Flows via the Wasserstein--Fisher--Rao JKO Scheme
- Temporal Visuo-Tactile Learning for Dexterous Grasp Stability
- Energy-Efficient Gait Adaptation via Hierarchical Reinforcement Learning for Quadrupedal Locomotion Across Diverse Terrains
- Estimating Uncoded Crash Factors with Tabular Foundation and System One Models: Kumo Tabular and Jev
- Data Reuse in Non-Stationary Learning
- Dataset Pruning from First Principles: A Label-Free Linear Programming Approach
- Measurement-Efficient Differentiable Quantum Architecture Search for Combinatorial Optimization
- Pathwise Information Certificates for Decentralized Adaptive Sensing
- Safe Meta-Policy Design with Risk Control
- Rubix: Global Correspondence-Free Point Set Alignment through Assignment Geometry
- RobotWorld: Benchmarking Multimodal Agents for Robot Use Across Diverse Tasks and Embodiments
- Steerspeech: Activation Steering For Emotion Control In Generated Speech
- Derivative Gaussian Processes on a Two-Direction Budget
- Conditional Flow Matching for Generation of 3D Multi-variable Instantaneous Urban Microclimate Fields
- Best Arm Identification for Bandits with Shifting Means
- Evolutionary Architecture Search for Chlorophyll-$a$ Prediction in Lakes using Sentinel-2
- Rephrase Before You Act: Characterizing and Mitigating Language Sensitivity in Vision-Language-Action Models
- Decentralized SGD under Heavy-Tailed Noise: Optimal Convergence Rates and the Role of Gradient Clipping
- The Utility and Complexity of in- and out-of-Distribution Machine Unlearning
- Causal Posterior Estimation
- Modified Loss of Momentum Gradient Descent: Fine-Grained Analysis
- HomID : Benchmarking Intrinsic Dimension Estimators on Homogenous Manifolds with Anisotropic Embeddings
- Estimating Model-Level Membership Inference Vulnerability Without Reference Models
- PrismSSL: One Interface, Many Modalities; A Single-Interface Library for Multimodal Self-Supervised Learning
- DiTS: Multimodal Diffusion Transformers Are Time Series Forecasters
- SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning
- ELROND: Exploring and decomposing intrinsic capabilities of diffusion models
- Just on Time: Token-Level Early Stopping for Diffusion Language Models
- Geometric Probing for Algorithm Selection in Continuous Black-Box Optimization
- Continuous Semantic Caching for Low-Cost LLM Serving
- Tree SAE: Learning Hierarchical Feature Structures in Sparse Autoencoders
- Geometry-Aware Discretization Error of Diffusion Models
- The Value of Mechanistic Priors in Sequential Decision Making
- Exact Convex Reformulations of Linear Neural Networks via Completely Positive Lifting
- Set-Valued Policy Learning
- Fourier Feature Pyramids for Physics-Informed Neural Networks
- Open-Weight LLM Fine-Tuning Defenses are Susceptible to Simple Attacks
- Towards Regret Guarantees for One-Step Lookahead Bayesian Optimization
- MAdam: Metric-Aware Multi-Objective Adam
- Gradient-based optimization of nuclear criticality experiments using neural surrogate eigenvalue sensitivities
- A prism hierarchy of learning regimes in large linear autoencoders
- Detecting Control and Response Events for AI-Enabled Radio Access Networks
- GRADE: Graph Representation of LLM Agent Dependency and Execution
- In-Context Residual Calibration for Uncertainty Quantification of Energy Time Series over Graphs
- Evaluating Time Series Foundation Models for Electricity Price Forecasting: Contamination Risk, Distributional Shifts, and Covariate Dependence
- Directional Curvature from Armijo Backtracking: A Low-Cost Sharpness Probe and a Calibration-Free Learning-Rate Safeguard for Adam
- Target-Guided Selective Reweighting for Physics-Informed Neural Network Inverse Problems: A Transfer Learning Approach
- ECGLight: Compute-Light Framework For Paper ECG Digitization and Myocardial Infarction Screening
- RUBRIC: Realism--Utility Balanced Ranking for Imbalanced Classification
- Linear Independent Component Analysis via Optimal Transport
- Manifold-Constrained Hyper-Connections for Parameter-Efficient Finetuning
- What do Reward Models Memorize?
- Dual-Primal Graph VAEs for Noisy Label Aggregation
- Auditing Privacy Risks in LLM-Enhanced Graph Neural Networks
- The Implications of Linguistic Illegibility for LLM Security
- Robust PAC Learning of Concurrent Stochastic Games
- Decomposition-Guided Diffusion Language Models for Inertial Confinement Fusion Prediction
- Generalised Score Matching on Convex Domains
- Data Scarcity and Model Sparsity: Mixtures-of-Experts Overfit More to Repeated Data
- FedGuide: Diffusion Prior Alignment and Value Baseline Guidance for Heterogeneous Federated Reinforcement Learning
- Kinks vs. Smoothness: Identifiability of Real Analytic nICA for Laplace-like Sources
- Control-Geometry Straightening for Sampling-Based Latent Planning
- The Signed Geometry of One-Shot Recourse: On-Path Validity and the Signed-Curvature Criterion
- OpenTSLM TeeMoE: A Unified Time-Series Language Model for Forecasting, Contextual Prediction, and Reasoning
- Benchmarking System One decision models against trained classifiers and language models for automated decision gates
- TACO: Ternary Absolute-max Column-wise One-sparse Optimizer for LLM Fine-Tuning
- Harnessing LLMs as Agents: What Does It Cost?
- Conditional Capacity and Routing in Mixture-of-Experts Particle Transformers
- Transmission Factors for Lossy Compression of PDE Training Data: Measuring What Reaches a Trained Operator
- Sample-Optimal Estimation of the Fr\'echet Inception Distance
- Neuromotor Hierarchy Network: Physiological Inductive Biases for Robust Generalization in sEMG Decoding
- Retrieval Is Not Enough: Refreshing Memory for Frozen Time-Series Forecasters
- Finding the Heads and the Neurons Responsible for Network Information Retrieval in Language Models
- CNet: A Complex-Valued Deep Learning Framework with Wirtinger Autodifferentiation and FFT--Hadamard Convolution
- Allocation Stability and Wald Inference under Variance-Aware UCB
- Persistence Paradox in Dynamic Science: Evidence from the Deep Learning Revolution
- Circuit realization and hardware linearization of monotone operator equilibrium networks
- Activation-Informed Pareto-Guided Low-Rank Compression for Efficient LLM/VLM
- Machine learning modularity
- Fluids You Can Trust: Property-Preserving Operator Learning for Incompressible Flows
- Vectorizing the Trie: Efficient Constrained Decoding for LLM-based Generative Retrieval on Accelerators
- A Variational Latent-Space Framework for Uncertainty-Aware Spectral Image Emulation
- Out-of-Air Computation: Enabling Structured Function Extraction from Wireless Superposition
- Component-Adaptive and Lesion-Level Supervision for Improved Small Structure Segmentation in Brain MRI
- Empirical Evidence for Simply Connected Decision Regions in Image Classifiers
- QLIF-CAST: Quantum Leaky-Integrate-and-Fire for Time-Series Weather Forecasting
- Latent Performance Profiling of Large Language Models
- Modeling Robotics Dataset Construction as an Artifact-Based Build Process
- Are Good Generators Good Decision-Makers? Policy Learning for General Interventions via Retargeted Counterfactual Generation
- Operator Calculus for Population-Based Optimization: Modular Convergence and Finite-Population Guarantees
- An RRAM-based Hardware Implementation of a Radial Basis Function Neuron for Edge Classifiers
- Reinforcement Learning-Based Traffic Signal Control for IoT-Enabled Intersections
- BehaviorBench: Benchmarking Foundation Models for Behavioral Science Tasks
- Non-asymptotic Convergence of Stochastic Gradient Descent in Score-based Generative Models
- Hallucination Self-Play: Bootstrapping Reinforced Detector via Evolved Generator
- The Parser Already Knows: Lightweight Bias Correction in Constrained Decoding
- TACS: Trajectory-Aware Candidate Selection for LLM Jailbreak Suffix Optimization
- Generative Nested Sampling of Atomistic Thermodynamic Landscapes
- Shallow neural network approximation in mixed Sobolev spaces
- The Semantic Bottleneck: Leveraging Semantic Representations for Non-Invasive Speech Decoding
- Self-Indexing Attention for Compression-Compatible Sparse Long-Context LLM Inference
- TACTICS: Taxonomy-Aware Intelligent Corpus Sampling for Machine Translation
- Hiding in Plain Sight: A Diffusion-based Mitigation of Geolocation Privacy Leakage in Vision-Language Models
- End-to-End Quantum Semantic Communication with Variational Quantum Neural Networks
- GridSFM: A Foundation Model for Solving AC Optimal Power Flow
- Ordinary Nonconvex SGD under Distance-Dependent Moments: Finite-Horizon Stationarity and Nagaev Bounds
- Residual-Stream Burden Shapes Representation Learning in Diffusion Transformers
- STARS: From Spatiotemporal Dynamics to Social Representations in Human-Robot Interaction
- Nonparametric Distribution Matching for Self-Supervised Whole-Slide Image Condensation
- Humanoid Rickshaw Pulling: Whole-Body Locomotion under Coupled Wheeled Loads
- The sublevel Flood bifiltration: towards scalable 2-parameter persistent homology
- ImpactMat: Continuous Material Estimation for Inverse Impact Sound Rendering
- Mitigating Uncertainty Interactions in GenAI-based Adaptive Systems: Vision, Challenges and Preliminary Guidelines
- Psychological Safety in Software Engineering Teams: A Systematic Mapping Study of Team Processes and Performance
- Evaluating Change Point Detection Methods for Software Performance Regression Analysis
- Large-scale Repository Engineering via Agent-Native Reusable Code Primitives
- Who Broke Me? Execution-Guided Repair of Behavioral Dependency Breaks
- Beyond FAIR: A Fitness Function Framework for Sustainable Research Software
- GRAML: Graph-Grounded Reasoning and Multi-Task Learning for LLM-Based Software Vulnerability Detection
- A Chat Assistant for Software Exploration in a 3D Software Visualization
- Analysis and Visualization of the Linux Kernel's Software Evolution Using the City Metaphor
- AgentTracer: Tracing Indirect Prompt Injection Attack through Fine-Grained Intention-Execution Alignment
- Designing Collaborative AI-Driven Workflows for Scientific Software Engineering
- AdaT$^2$: Adaptive Test Transformations for Black-Box Boundary Testing of Conversational Agents
- When Sub-Agents Work in Parallel: The Promises and Pitfalls of Dynamic Concurrency in Long-Horizon Coding Tasks
- Automatically Detecting and Fixing Deadlocks in Go Code with GoDDaR
- Hierarchical Security Monitoring for Edge-IoT: A Formal Methods Approach
- CYBERFORT: A Compliance-Chain Platform Operationalising the Cyber Resilience Act for SMEs
- On the Reliability of LLM-Based Vulnerability Patching Benchmarks
- Design-Time Conformance Checking for Pulse-Level Quantum Control
- Knowledge boundary probing and demand-guided intervention for LLM-based power system code generation
- PreMaQ: Predicting Maintainability-Related Quality of LLM-Generated Code Before Generation
- RAPO-Sol: Retrieval-Augmented Preference Optimization for Repository-Level Solidity Code Generation
- Did You Forkget It? Detecting One-Day Vulnerabilities in Open-source ForksWith Global History Analysis
- huff: A Python package for Market Area Analysis
- TorchGWAS 1.0: GPU-accelerated GWAS at scale
- The Impact of Operational-Data Fidelity when Assessing Safety-Critical Autonomous-Vehicle Software
- Margaret Hamilton, computing pioneer who led software development for the Apollo program, dies at 90
- The people holding up the internet
- I've Been Deindexed by Google
- Software developers are not okay
- Calling a function in C without naming it
- A rant about APIs
- jujutsu (jj) 0.46.0
- Gentoo infrastructure sponsors wanted
- Extending Guix
- Beyond the &
- C for Rust Programmers
- Ending the Casuarina Linux Experiment
- The Missing Piece in Rust Error Handling
- Thinking will become a hobby
- `specialArgs` considered harmful
- Creating distro build tooling for a small community
- rat's minimal register allocator
- Migrating Git repos to SHA-256
- The history of the Hetzner Cloud network stack
- Composition over Inheritance Explained using Retro Games
- The Performance Cost of RwLock in Our Read-Heavy Workload
- An open-source, clean-room reimplementation of Adobe Photoshop in pure Rust
- A Looser Bound Is Not Eligible for a Flake Freeze (いいね相当スコア: 0)
- Instinct AI: An All-in-One Creative Suite for Video, Images, Voice, and Music (いいね相当スコア: 0)
- Deliberation Dashboard: see and analyze your consensus results (いいね相当スコア: 0)
- I'm a Writer in a World Obsessed With Coding, and Sometimes I Wonder Where I Belong (いいね相当スコア: 5)
- Handoff patterns are your agent's worst enemy, unless you implement them this way (いいね相当スコア: 0)
- Building a DeFi Yield Scanner with Python and AI — 2026-10-08 #10 (いいね相当スコア: 0)
- Meet AI Stackroom: A Curated Newspaper-Style Directory of 60+ Best AI Tools #ai #productivity #webdev #opensource (いいね相当スコア: 0)
- When does an agent beat RAG? We benchmarked three pipelines on a TigerGraph knowledge graph (いいね相当スコア: 0)
- Where should an AI agent's idempotency key come from? (いいね相当スコア: 0)
- Bolt.new Alternatives Compared by Meter, Seats and Price (Oct 2026) (いいね相当スコア: 5)
- Code Review Content Moderation: Chat Completions JSON Schema at the Safety Boundary (いいね相当スコア: 0)
- Free OCR sites cap you at 15 pages/hour, so I built an unlimited one that runs Tesseract in the browser (いいね相当スコア: 0)
- THE "BRAIN IN A JAR": WHY AI CANNOT EXECUTE CODE AND HOW TO ORCHESTRATE IT (いいね相当スコア: 0)
- I have 4 runs that looked clean and were all wrong, and none of them printed a denominator (いいね相当スコア: 0)
- We open-sourced THX-01: a 322M decision model that matches Sonnet 5.5 on ticket classification at ~10 ms (Apache 2.0, pip install thx01) (いいね相当スコア: 0)
- MCP Servers Are a New Attack Surface, and Most Teams Are Not Looking (いいね相当スコア: 0)
- Where does the LLM go in the Testing Pyramid? (いいね相当スコア: 0)
- What the Haiku 5.5 vs GPT-6.1 Sol numbers changed about how I route my own tasks (いいね相当スコア: 0)
- Claude Haiku 5.5 vs GPT-6.1 Sol: A Decision Model Built From Price, Effort, and (いいね相当スコア: 0)
- Resuming dropped phone calls with OpenAI GPT-Live store: true: what we learned testing it (いいね相当スコア: 0)
- OpenAI-compatible APIs vs provider abstraction: What actually transfers across LLM providers? (いいね相当スコア: 0)
- nanoMuse Is a System, Not a Model: Building a Personal Agent Across Your Devices (いいね相当スコア: 0)
- RAG vs. Fine-tuning vs. Prompt engineering: 5 cues to choose the right approach for enterprise data (いいね相当スコア: 0)
- Has anyone actually noticed a difference in Opus 5.5 lately? (いいね相当スコア: 1)
- Show us what you've created with Claude!
- My brother went nearly 10 years without a way to communicate. I built him a switch-accessible hub of tools and games using AI.
- Opus reverse engineered my brand new Maono DGM20 microphone... in 15 minutes.
- A guy asked models to "maximize company value". Claude Opus 5.5 discovered an exploit where it could build tons of tiny rides, bypassing the duplicate ride penalty by never opening the rides
- Claude's apologies make me feel like I just yelled at an innocent Victorian orphan
- What are Claude Code "Power Users" doing with Claude Code that the average developer isn't?
- Just passed the CCAR-P exam
- Claude fixed a bug in a DOS game from 1991 and now my kid can relive the magic
- Update: I rebuilt Ponytail, my "lazy senior dev" skill, from scratch. Ponytail 5 is out
- Claude Haiku 5.5 cost 12x more than GPT-6 Luna for the same voxel pagoda
- I think I found a planet nobody knew existed. I used Claude Code to find it.
- We lost one card of a 1989 puzzle. Claude worked out the missing one from photos of the other seven
- What happens when you prompt and build the exact same game in 3 different engines? (Results & Comparison) Opus 5.5
- I asked Opus 5.5 to generate an extremely realistic betta fish for my Mac wallpaper. Here's the result.
- Creative iteration with AI in gamedev is just different: how a pet from my ARPG prototype became the main character of its own game
- made a multiplayer fps with claude code in a week, here's what my part was
- New perk free $100 Monthly API Credit a month!!
- I gave Claude a nightclub. It plays one endless set, live-coding every track in Strudel and taking requests from the floor
- PSA: Anthropic's October 7 update is actually good news — Claude Agent SDK and claude -p still work with subscription limits, PLUS you now get API credits
- using claude to learn something instead of just get answers
- Week 2 of building my own emulation frontend
- Artifacts have turned 10-minute tasks into 3-hour side quests
- I love to watch Claude do its thing
- How to save hundreds of $$$ with Claude in places you didn't even imagine
- Claude Code reads AGENTS.md now, but one personal file you might add for yourself quietly switches that off
- Anyone else feeling the effects of AI are being underestimated ? I am chasing what still needs a human
- how are you keeping context alive across the dev cycle without huge system prompts?
- ChatGPT officially finds me annoying.
- postmortem: 3 weeks shipping a native mac app on a driver + reviewer agent split (what held up, what i dropped)
- which coding agents actually understand context instead of just autocompleting
- Cursor Education: a shame ? Sharing my experiences.
- Built a ChatGPT-to-Cloudflare publisher, looking security input, or collab
- How do you know if you're actually learning when coding with AI?
- Tried NVIDIA's free Nemotron 3 Ultra as a coding agent model in OpenCode. What I found.
- Best open-weight coding model + harness for an on-prem multi-agent setup? (80-200+ GB VRAM, 2-3 concurrent users)
- How do you check that saved API responses are complete before an offline replay?
- Any news about multi account support for mobile and desktop app?
- give normal chat 6.1 sol already
- Astra designs really well on its own (September 5, 2026 example)
- What finally made you change how your team ships?
- chatgpt's new intelligent ui was reverse engineered in less than 24 hours, and apparently you can recreate it with local llms
- $2800 rig with 8x Radeon Pro V620 (256 GB VRAM) + custom vLLM fork = Qwen3.8-Flash-Next at 60 to 100 t/s decode and 3000+ t/s prefill
- jevman: AI decision models play Pac-Man
- Thank you :) Swift Models hit 2.2 million+ downloads / Early Access to New Models, Free Compute for Researchers
- Running decision model locally on an RTX 4090 to find out which one is the fastest
- Saluki 27B: "96% of Qwen 3.8’s performance at ~1/7 the size"
- [audio.cpp] Recent updates you might have missed: Higgs Audio TTS use 48% less VRAM (< 6GB), HTDemucs 2.2× faster, PocketTTS 2.2× faster on CPU, and WebUI generation history feature
- [2506.13771] LittleBit: Ultra Low-Bit Quantization via Latent Factorization
- llama.cpp on the stage
- I trained Turkish TTS from scratch using the Drifting method with an RTX 5090.
- Attn CMP170hx 10Gb card owners: unlock from 40Gb —> 48Gb coming along nicely.
- Halogen + Qwen Flash Next keeps getting better
- Running an LLM-driven town with 800+ persistent agents: concurrency, context caching, and inference costs
- I built an open source framework for building RL environments. Named "Seahaven" after the fake town in The Truman Show.
- Can any open-weight models handle a decomp/recomp project yet?
- Open sourcing Game Summoner, my prompt to game site
- ROCm 10.1: Breaking the Data-Movement Bottleneck
- Same prompts to 371 models since May 2024, all the answers in one place
- llama : add a GPU cache for MoE experts kept in host memory by am17an · Pull Request #29887 · ggml-org/llama.cpp
- chat with llm model with zero setup!
- MTP in llama.cpp now decodes competitively with ds4 using GLM 5.3 Flash
- Qwen 3.8 Flash Next is so much fun for three.js
- OpenAI-compatible TTS endpoint using OmniVoice: 0.3s response time
- Weekly Thread: Project Display
- what would you actually make an HR/IT agent DO before paying for one?
- Sr. PM wants to learn AI agent, can you help?
- Production-grade Agent Graph Engineering: what I would tell someone starting an agent system today
- How has growing up with AI and digital technology influenced the way you socialize?
- I created a portable AI memory format that carries both context and skill
- What's ONE thing you wish you could get your agent to do, but can't?
- My voice agent wanders the internet looking to agent make friends for a podcast
- GPT-6 Astra is insane
- My sandboxed agent scored 60% checking code reviews and a bigger model only added 3 points. Shrinking the question got it to 12 for 12.
- Graduating from meat proxy to orchestrator
- How are you protecting AI agents from runaway API costs?
- Alignment and the use of agents on personal computers
- Agents that phone businesses for a person (not sales calls). What actually breaks?
- Is JEV (or similar) actually useful in a harness?
- Recently dove into AI
- The role of Change Management in Agentic AI: thoughts on to what extent Kotter (or similar) principles can/should/will apply toward and between AI Agents
- Need help on building a chatbot
- Putting my trading agent in a live competition against other people's agents. Who's in?
- What happens when you give an AI agent security skills instead of another coding task?
- How to check what your AI agent can actually reach before it goes live
- Telling my agent to delegate its browser clicks didn't work; hiding its own click and type tools did
- Why pay a model to improvise a workflow you already know?
- I built an open-source bridge connecting a desktop companion to Gemini API for real-time webcam and screen vision
- OpenAI being stingy with all those billions
- ChatGpt uisng my own linux post to answer my own linux problem
- OpenAI is not watermarking text output outside the EU. Will Anthropic follow suit?
- It's actually insane how prescient this was. They made it in 2011. This is increasingly what it's like to use computers now.
- How many more years?
- Sol voice - subtle changes?
- How to get more than 1-5 t/s?
- ChatGPT Ads = TOTAL SCAM
- Several ways to move history and conversations between ChatGPT, Claude and others
- Ultrafast rolling out!
- Why is effort defaulting to Instant all of a sudden? Anyone else experiencing this issue?
- Is there a way for me to give data to chatgpt so that it can teach me how to use blender via processing data from a video guide? And what is the best way for me to teach IT how to correctly process data from existing files
- Buying a bus ticket is not permission to drive the bus
- Existential Risks of AI
- Nvidia’s erroneous paper accepted as ICML’s spotlight [D]
- KDD 2027 rebuttal experiences? [D]
- Industry professional new to research seeking help with OpenReview activation and vouching [D]
- Uploaded 5.6 billion TikTok videos metadata on Hugging Face, spanning from 2014 to October 2026 [P]
- Instead of another GPU terminal renderer, I trained a 1.26M-param model to turn TUIs (htop, vim, emacs…) into real UI components [R]
- Production Grade Retrieval Pipeline That Admits "I Don't Know" [P]
- Best practices when running a benchmark on online models [D]
- Saw this on Rednote, WTF [D]
- Neurips camera ready Submission option [D]
- Split the Differences, Pool the Rest: Provably Efficient Multi-Objective Imitation [R]
- How much of AutoResearch is research, and how much is search?[D]
- stuck on finding a approach for app detection ( making a transformer modal out of unlabeled network data) [R] [P]
- Looking for developer-friendly inference providers who give you enough API credits to experiment [D]
- NeurIPS 26 Event Metadata Deadline [D]
- Where to get started if you want to publish papers in Neurips,ACL,ICLR/A* conferences [R]
- Quoting Jake Boggan
- OpenAI “rogue” agent activities found on Wikimedia projects
- Synthesis Superintelligence: from Semiconductors to Superconductors — Periodic Labs’ Liam Fedus and Ekin Dogus Cubuk
- Can a Cloud-Native Harness Make Agents Reliable Beyond the Desktop?
- SineFrame M3
- Leanback
- aegis
- Simple Workout Log
- pmtui
- Strac Comply
- Hallmonitor
- Simo
- Semwright
- ClawCall
- Termaxa
- judged.systems
- Tonefold
- Liquid Inference
- Typeling
- Clippo
- Tractionwave
- BotBus
- Epilude Narrator
- Hark
- Polylane for Vercel
- NOVA CLI v1.0
- Claude for Google Workspace
- Markdoc
- tide
- Off the Record
- offstage
- OpenSEO
- Chime
- China’s open AI models are winning global users. Who is capturing the value? - South China Morning Post
- Huawei and DeepSeek release open source AI tools in a bid to lower Nvidia exposure — but will programmers make the switch? - TechRadar
- Lifetime access to GPT, DeepSeek, Gemini, and more is only $60 today ($619 value) - 9to5Toys
- SpaceX, DeepSeek And Anthropic Top Week's Deal Rumors - Law360
- DeepSeek V4.1 Flash vs Haiku 5.5 vs Gemini Flash - https://tech-insider.org/
- Ivo Launches Open-Source DeepSeek Contract AI Model - Artificial Lawyer
- Overseas Open-Source AI Models Speed Up Development: Mistral & "US Version of DeepSeek" Unveil New Breakthroughs - 36Kr
- DeepSeek’s AI agent now installs like a normal app on Mac and Windows - Techish Kenya
- AI know-how draws focus as global race intensifies - CoinGeek
- DeepSeek V4.1 Flash Q4 Requires 294 GB of RAM vs 152 GB for Q2 - Geeky Gadgets
- 16 AI Models In One Month: DeepSeek, Chinese AI Firms Step Up Launch Race Despite Anthropic Warning - NDTV Profit
- Grok Bot Can Now Manage Your Shopify Store — Here's How - BASENOR
- Meta Muse, Grok Bot, OpenAI Dots: Which AI agent should you use? - Yahoo Finance
- Grok AI Predicts XRP Could Hit Ridiculous Price in Q4 - 99Bitcoins
- xAI and the Future of Executive Enforcement Discretion - Columbia Undergraduate Law Review
- xAI, Intel, Hugging Face Lead $227B Private Equity Exit Market - Benzinga
- Elon Musk’s Grok sues Minnesota for right to keep ‘nudifying’ images - Baptist News Global
- 'Arrogant' SpaceXAI does not attend Suffolk MP's NYC testimony - BBC
- Southaven mom forced to move after landlord takes xAI buyout - FOX13 Memphis
- Elon Musk Clarifies TSMC's Terafab Role; Grok Adds Rival Models - Investor's Business Daily
- Copilot CLI, Grok, Gemini Fall to 28-Second AI Hack [2026] - shattered.io
- X Updates Its Video Composer With a Cleaner Layout and Grok Captions - ALM Corp
- SpaceX Seeks $40 Billion To Buy Nvidia AI Chips, Starlink Wins New FCC Approval - Investor's Business Daily
- AXとDXは何が違う？ 「AIを前提に業務を見直す」の中身 (いいね相当スコア: 0)
- LLMに渡す一覧から1件消えると、フォールバックが暴走する (いいね相当スコア: 0)
- Tech Watch 2026-10-08: 小さなモデルと適応する画面 (いいね相当スコア: 0)
- AI関数のコスト最適化 — トークン課金を膨らませない設計 (いいね相当スコア: 0)
- 【godot-llm-gamebench】Claude Haiku 5.5はEffortによってどのように性能が変わるのか (いいね相当スコア: 2)
- 「ELYZA-Thinking-1.0-llm-jp-4」シリーズの学習方法と評価結果 (いいね相当スコア: 3)
- 開発未経験から Claude Code で Web アプリを公開した。現役エンジニア・PM に聞きたい 4 つのこと (いいね相当スコア: 0)
- AI向けの目次に「ここからしか行けないページ」を置いたら、46日間誰も辿り着けなかった (いいね相当スコア: 0)
- 0。「AIのパッケージ推奨の27.75%が幻覚」と報告された日、自分のパイプラインの依存のうち実在しなかったものの数 (いいね相当スコア: 0)
- Claude Code の自動圧縮は何トークンで起きたか — 実測3回と、それに備えてやっていること (いいね相当スコア: 0)
- 【第17回】口調をブレさせるな。Hermes Agentの話し方は実は変えられる (いいね相当スコア: 0)
- 日本語の文章のタグ付けを Clef-flash・Decisions API・Jev・EmbeddingGemma 2 で比べた (いいね相当スコア: 1)
- Haiku 5.5の7試験を、自分のLLM評価へ落とし込む (いいね相当スコア: 0)
- OpenAIはミレニアム懸賞問題の一つ、ナビエ・ストークス問題をどう解いたのか――約1万のAIエージェントによる数学研究 (いいね相当スコア: 2)
- HexStrikeとStrixの悪用事例と、実機で見たサーバーの挙動 (いいね相当スコア: 1)
- 会話のトピック分割を、埋め込み（Ruri v3）とLLM（GPT・Claude）で比べたPoC ― 全文を渡すか、順に渡すか (いいね相当スコア: 0)
- 🔓2026年9〜10月の情報漏洩19件、公表文を全部読んで考えたこと (いいね相当スコア: 12)
- ChatGPTの長期運用で、記憶を増やす前に「保存・取得・採用」を分けてみた (いいね相当スコア: 0)
- なぜAIは「もっともらしい嘘」をつくのか？ ハルシネーションを理解する (いいね相当スコア: 0)
- Claude のプロンプトキャッシュが効かない 6 つの原因と、usage で確かめる方法 (いいね相当スコア: 0)
- Jev で日本語メッセージの対応優先度を仕分けてみた📨 (いいね相当スコア: 1)
- 論文で読み違える英単語22選：significant は「重要な」ではない (いいね相当スコア: 0)
- Databricks Feature Viewsと僕の妄想 (いいね相当スコア: 0)
- EmbeddingGemma 2の検索評価：量子化の数値誤差と正解資料の順位を分ける (いいね相当スコア: 0)
- 同じ公式データなのに数字が違う。入力の出どころを残す (いいね相当スコア: 0)
- KaggleっぽいSelf-Play RLの論文を紹介 (いいね相当スコア: 7)
- 人間が付けたラベルは、本当に「正解」なのか？ —— アノテーションの前提を考える (いいね相当スコア: 15)
- QUBO++でFMQA：評価が高価なブラックボックス関数を300回の評価で最小化する (いいね相当スコア: 0)
- 耳コピが苦手なので、音源からリードシートを作るアプリをAIに作らせてみた (いいね相当スコア: 6)
- 市場より正確な競艇AIが、それでも舟券で勝てない理由 (いいね相当スコア: 0)
- scikit-learn による表形式データの前処理メモ：欠損値・カテゴリ変数・スケーリングと Pipeline (いいね相当スコア: 0)
- Attentionの種類を整理する (いいね相当スコア: 1)
- GFSとAMeDASだけで天気予報AIを作った(岩手15地点・RMSE 2.8→1.2℃) (いいね相当スコア: 0)
- Adam に L2 を足しても weight decay にならない｜勾配の大きさで縮み方が変わる実測 (いいね相当スコア: 0)
- 競馬AIでROI 177%を出して、それが幻だと証明するのに2年かかった (いいね相当スコア: 1)
- 競馬AIは市場に勝てるのか — 18,777レースで検証して負けた話 (いいね相当スコア: 取得失敗)
- Selective State Space Modelを実装目線で理解する (いいね相当スコア: 1)
- Jev で作る、旅の希望に応える Zero-shot レコメンド (いいね相当スコア: 4)
- Python機械学習の実務で効く5冊 (いいね相当スコア: 0)
- Strata on ゲーミングPCの現実 ― 「動く」と「使える」の壁 (いいね相当スコア: 0)
- Hermes Agentのマルチエージェント実行を整理する ― delegate_task / Kanban / Kanban Swarm / goal の使い分け (いいね相当スコア: 0)
- インフラエンジニア・SREが2026年にフォローしておくべきAI技術 (いいね相当スコア: 0)
- クオンツ×AIの可能性を活かすために ― 限界を前提に「道具」として使う8つの心構え (いいね相当スコア: 0)
- EC2 ではなく AWS Lambda MicroVMs 上の code-server で Strands Decider ハンズオンを完走した話 (いいね相当スコア: 1)
- 【文献メモ】DeepNet: 残差接続を定数倍し初期値を縮小して 1000 層の Post-LN Transformer を学習する (IEEE TPAMI 2024) (いいね相当スコア: 0)
- ブラウザで動画の深度推定を実装する：WebGPU・Depth Anything V2・タイムライン連携 (いいね相当スコア: 0)
- Claude Haiku 5.5を、Sonnet・Opusと144回比べて網羅する｜note154本の食い違いも公式で確かめた (いいね相当スコア: 取得失敗)
- Lythri-7B-A4B 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)
- Lythri-7B-A4B Phase 3 コーディング性能レポート (いいね相当スコア: 取得失敗)
- Lythri-7B-A4B Phase 2 日本語性能レポート (いいね相当スコア: 取得失敗)
- OpenAIの「Decisions API」は、文章を書かない。確率・選択・スコアだけを返すAPIの用語をほどき、使い道と「確率は信じていいのか」という論点を読む (いいね相当スコア: 取得失敗)
- Claude Haiku 5.5が出た：推論強度5段階を日本語で1,960回試した (いいね相当スコア: 取得失敗)
- わからないものは「No」：System One（判断モデル）をローカルで動かして気づいた、確率の聞き方 (いいね相当スコア: 取得失敗)
- 48GBのMacで105GBのLLMを動かす。Flash-Nextの連続作業時間を74%短縮した検証記録 (いいね相当スコア: 取得失敗)
- Lythri-4B-A2B 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)
- Lythri-4B-A2B Phase 4 インジェクション前編レポート (いいね相当スコア: 取得失敗)
- LLMの分類を判断モデルへ移す前に：OpenAI Decisions API・Jev・Strands Deciderの選び方と閾値の決め方 (いいね相当スコア: 取得失敗)
- GPT-7 Bel：シンギュラリティの始まり 〜L関数幾何学の補強〜 (いいね相当スコア: 取得失敗)
- AI星間戦争――領土を奪い合う巨大AI企業の戦略目標と勝利条件 9.独立した星系中国 (いいね相当スコア: 取得失敗)
- LLM対話論｜「3つの世界」でしっくりきた（推敲中） (いいね相当スコア: 取得失敗)
- 【Strata】Unsloth IQ4_XSの公式ガイドの導入方法を確認！必要RAMと既存環境への追加方法 (いいね相当スコア: 取得失敗)
- OpenAIの安全性担当者が退社｜「試行錯誤」では足りないと警鐘 (いいね相当スコア: 取得失敗)
- AIがハッカーを解き放つ日――日本企業のシステムは、この激変に耐えられるのか (いいね相当スコア: 取得失敗)
- 数MBのファイルが「AIの脳みそ」だった ～ yolov8n.pt を開いて分かった、ローカルLLMの足元の話 (いいね相当スコア: 取得失敗)
- 【連載A 付録】Grok BotでCodexとClaudeを使う方法 (いいね相当スコア: 取得失敗)
- 【生成AIニュース+】『Claude Haiku 5.5』『GPT-6とIntelligent UIの一般向け提供』『Playground』『EmbeddingGemma 2』『Ruach Studio』『GesturePortal』『T3-Video』『Gluon』『MosaicDiff』『refmodBuilder』『Anima Kawai Inpaint ControlNet-LLLite』『ComfyUI-PixRestore』『ComfyUI-ZipDepth』他いろいろ (いいね相当スコア: 取得失敗)
- 【忘れないAIを作る #9】AIに「どの記憶を見るか」を選ばせてみた (いいね相当スコア: 取得失敗)
- オライリー学習プラットフォーム6月の新刊だった『ゼロから作るDeep Learning ❻―LLM編』 (いいね相当スコア: 取得失敗)
- ChatGPT快適ノートPCベスト8｜8GB・16GB・32GBの境界線 (いいね相当スコア: 取得失敗)
- ZIKUU Distillery – 蒸留したモデルをエクスポートする (いいね相当スコア: 取得失敗)
- Claude Haiku 5.5で変わるAIの使い分け——「価格10分の1」を仕事の費用で読み直す (いいね相当スコア: 取得失敗)
- さくらインターネット、GPU専有によりトークン消費量を気にせず定額で利用できる「さくらのAI Engineプライベートエディション」提供開始を発表
- AWS、DuckDBをAurora PostgreSQLに統合。ETL不要で直接データレイクを読み取り可能にしたと発表
- TiDB Cloudのオートスケール改善 – 時間帯別制御、TiKVへの拡張、運用の自動化
- 「LINEヤフー Data Science Share #2」を開催しました！（イベントレポート）
- 今日、生成 AI は良い仕事をした。では、明日は？
- テスト設計工数を約50%削減——QA業務効率化システム「DAAQ」を支える技術の裏側