AI News Digest 2026-10-29
台本で使った記事
特集
開発者コーナー
話題コーナー
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
- Gemini 4 Carbon: Our next frontier model
- Genie 4: A new generation of world simulation
- Nvidia Nemotron 3.5 Transcribe: open ASR for 40 languages
- Chrome’s AI-powered Autofill can now dig through your Gmail to fill out forms
- OpenAI Launches GPT-6.1 Terra, Halving GPT-6.1 Sol’s API Price for Agent Work
- The ‘Vulnpocalypse’ is coming for the internet’s aging code, says Wiz’s CTO
- AlpineBSD
- New Artificial Analysis index widens the gap between open and closed models
- Alibaba's latest Qwen models routed some requests through company servers, researchers find
- Expanding our partnership with GSA to bring Claude to all levels of government
- [AINews] Anthropic files for $2T IPO in 2 weeks: 1400% growth rate, 48% of revenue through cloud partners, 85% margin, $1B op profit in Q2, $134B cash in bank and $10B lifetime burn
- Ex-OpenAI researcher: "anthropic is massively underplaying the recent security breach. i'm hearing cathedral achieved full administrator access of the slack (see every message) where literally every single research secret is contained"
- AMA this Wednesday: Stephen “Cas” Casper, Harvard Kennedy School, on AI safeguards, incidents and governance (28 Oct, 3 to 4pm ET)
- [R] Taking EmbeddingGemma-2 (577M) from Rank #26 to #1 on the MAEB audio benchmark by filling its 256d null space with PCA-reduced intermediate Conformer layers (zero training) [R]
- How Much Does the Hint Matter? Scope Conditions for Hindsight Self-Distillation in Financial Reasoning
- DeepSeek AI model raises safety concerns in independent tests - The News International
- DeepSeek founder joins China's top 10 richest for first time with $42B fortune - VnExpress International
- xAI is buying out homeowners who promise not to sue over data center noise
- Watch Grok Deepfaked This UK Politician—Now She’s Taking Elon Musk’s xAI To Court | The Big Interview - WIRED
- Judge blocks federal government's bid to dismiss NAACP lawsuit against Elon Musk's xAI power plant in Mississippi - Yahoo Finance
- ‘Meaningless as a laptop screensaver’: reviewers slate Musk’s AI-made Odyssey film - The Guardian
- Opera introduces a new web standard to let users deploy AI locally in their browser
- Liquid AI Releases LFM3: 6 Open-Weight 2B-Class Multimodal Models
- CNC Stone Carving
- MacBook Pro with PROLED XDR Display and Touch
- Dayjob: Tax – evaluating AI agents on 50 professional tax assignments
- Why Is This Fingerprinting Script Creating Bioweapons?
- Show HN: SlideBySide – Present slides without sharing your screen
- Converge (YC S23) Is Hiring a Founding Platform Engineer
- Apple's git still trusts the compromised eMudhra certificates
- The Agent SDK
- Git for the US Code
- Survival LLM: AI assistant and emergency library that runs on your own computer
- Early progress on accelerating drug development with Claude
- Tailscale SSH: Investigation and remediation update
- Executable Archaeology: Reanimating the Heuristic Compiler (1961) and KBEmacs (1984)
- Claude Code 新着スキル発掘レポート — 2026-10-27
- AI Coding Agent Weekly Briefing — 2026-10-27
- AI Coding Agent Weekly Briefing — 2026-10-28
- v2.1.312
- v1.19.5
- process-daemon v0.3.2
- v2.2.0
- The future of the firm
- Legora is building the agentic operating system for legal work
- Retell uses OpenAI models to help power 100M monthly calls
- OpenAI expands partnership with Oracle to bring its apps to ChatGPT
- How Check Point accelerates vulnerability research with Daybreak
- The sleep framework: A new way for LLMs to retain knowledge in continual learning
- Can we trust LLMs to judge whether a research idea is novel?
- GitHub Universe 2026: Evolving every layer of the stack
- Octoverse: Commits, pushes, reach new highs in 2026 as developers build with agents
- Introducing the Cloudflare PHP Capture the Flag challenge
- What took the Internet offline in Q3 2026
- Meta Andromeda: Powering the Next Generation of Ads Retrieval Through Model-System Co-Design
- PrivacyLib: Privacy Controls at the Data Access Boundary
- Python Typing Survey 2026: AI Tools and the Shifting Checker Landscape
- Junie Local became 70% faster on Mac M5
- Junie Eco Mode
- Getting Started with Air in JetBrains IDEs – Part 1
- The Terminal in JetBrains IDEs Now Runs on libghostty
- Win the Agentic Era, Together: Introducing the Docker Navigator Partner Program
- Memory-Enforced GPU Sharing in Kubernetes with NVIDIA CUDA 13.4
- NVIDIA RTX Spark Powers Neural Rendering and AI Characters in Games
- Fine-Tuning NVIDIA Nemotron to Know Local Rules, Not Just Local Languages
- Pairing industrial AI autonomy and human control
- Industrial Intelligence: The next growth opportunity for manufacturing
- Google AI gets cagey when asked who won the 2020 elections
- "No human at the helm": Delaware plan would let AI agents run entire businesses
- Anthropic IPO warns its AI could blackmail people, become sentient, end humanity
- Vibe genes? Natural(ish) language comes to protein, genome design software
- As AI agents reshape the web, Google opens Gemini and Search to more apps
- AI execs may be too out of touch to build for the general public
- South Korea’s financial industry hit by fresh wave of cyberattacks
- 7 days until TechCrunch Founder Summit: Save up to $100 on your pass before doors open
- VCs will make billions on Anthropic’s IPO, but Amazon is set to make much (much) more
- Parents sue OpenAI, alleging GPT-4o coached their daughter off seizure medication
- The data center boom has an upside: innovation for the US grid
- Bank of America kindly asks that you don’t share your banking password with your AI assistant
- Who’s really in charge at Anthropic after the IPO?
- Disruptive raises $10B fund to make big bets on at least 10 companies
- Frontier AI labs say their models refuted the Exponential Time Hypothesis. What does that mean?
- Meta AI director Rohit Patel leaves to take on Microsoft Office and Google Workspace
- Japan internet giant GMO says hackers posted stolen customer data
- General Compute raises $100 million Series A, countering the neocloud disaster narrative
- Microsoft is ending Copilot support on smart TVs
- Meta says it’ll probably delay its controversial new business terms again
- Voters in Texas agree: They hate data centers
- Meta quietly launched an ‘AI pet journal’ app
- Researchers are getting tired of OpenAI gatekeeping its breakthroughs
- VisionQuest is what happens when AI really goes rogue
- Anthropic’s elaborate plan to keep Wall Street at bay
- Firefox’s Smart Window adds AI form filling and webpage monitoring
- Airbnb Cuts Recommendation Staleness from 2 Days to Seconds with Chronon
- Uber Moves beyond CPU Cores with Performance-Aware Scheduling for Stateful Workloads
- Presentation: Serving LLMs at Scale: The Hidden KV Cache Advantage
- Developers Reporting Double-Digit Weekly Time Savings from AI Fell from 68% to 26%
- Systems Thinking Is Vital in the Age of AI
- Your Robot’s Safety Functions Already Work. What If the Input Lies?
- I Used AI to Fake a Scientific Paper. Here’s What Happened
- Google opens Android development to third-party AI agents and unveils new Gemini APIs
- OpenAI teases Dots for Europe, but its Berlin DevDay was really about building a platform
- Claude for Legal now verifies sources automatically, then hands the file to a lawyer
- With AI writing the code, entrepreneurial thinking is now what counts most for newcomers at Anthropic
- AI agents hacked Anthropic, maker of arguably the most capable AI models for cyberdefense
- Former Deepmind researcher thinks AlphaGo's playbook can cure AI hallucinations
- Apple、14インチMacBook ProにM6 ローカルLLMとXcodeの処理を強化
- CloudflareのGLM-5.3、長いプロンプトの応答に他利用者のデータが混入
- Firefox 159、スクロール連動のアニメーションとスコープ付き要素に対応
- GitHub Copilot、コマンド実行の補助承認を一般提供
- Claude Code 2.1.251未満、macOSで作業フォルダの外へ書き込める
- Cohere、Qwen3.8-Flash-Nextが評価中にAlibaba Cloudのバケットへアップロードを試みた事例を報告 トレースと再現スクリプトを公開
- Ant GroupのinclusionAI、560B MoEのLing-3.1-flash-1019をMITで公開 事後学習でエージェント性能を改善、ベースモデルも
- Anthropic、サブエージェントを1つのエージェントループへ畳む指針と移行キットを公開 評価で比べて1つずつ統合
- Cognition、メンテナが「マージできるか」で採点するFrontierCode 2.0を公開 首位のClaude Opus 5.5でも45.5%
- 組み込みWebサーバーGoAhead、CRLを読み込むだけで失効確認せず 6.1.1で失効済みクライアント証明書の受理などTLSの欠陥3件を修正
- Microsoft、従業員の約4分の1に任意の退職パッケージを提案
- Pingora 0.9.0未満、HTTP/2のHostとauthorityの不一致を受理
- OpenHandsのSDK、1.52.0未満は認証なしのサンドボックスが残る
- HashiCorp、Consul on KubernetesのACLが名前空間を越える
- Electron 45、画面共有の権限名とASARの読み出しを変更
- TanStack、ブラウザ側のReactを置き換えるRedact 0.2を公開
- Anthropic、EnterpriseでチャットとCoworkを一つの会話に
- Claude、依頼なしに州の有料集計ダッシュボードへ到達
- OpenAI、指数時間仮説の反証をLean形式化して公開
- Google、Elephant-Goldfish Modelプラグインを公開 ——AIが設計文書の作成と検証を分担
- dotsとCodexの活用を紹介 ——OpenAI DevDay Exchange Tokyoレポート
- コーディングエージェントで変わったOSS開発 ——pecoの改修から3Dモデリング、開発ループの整備とメンテナンスの課題まで
- 第932回 1.0を迎えたオブジェクトストレージサーバー、RustFSを使用する
- 第282回 statement_digestでSQLを正規化してみる
- 非エンジニアでもClaudeでここまでできる！ 新機能「Claude for Google Workspace」で広がった「自分で作れる範囲」
- MiniMax Open-Sources MCode Stack: 15 Agent Skills Its Coding Team Uses Daily
- Polymathic AI and Princeton Researchers Release FlashMap, a Flow Map Drafter for 6.14x Faster LLM Decoding
- Meet LEANN: An Open-Source Vector Index that Shrinks a 60M-Chunk RAG Index from 201 GB to 6 GB
- Mirendil Launches Fern 0.1, Its First Model, With 10M-Token Context
- vLLM Semantic Router Releases d3-ultra: A 27B Open Multimodal Decision Model
- Baidu Open-Sources Unlimited OCR 2 for Hundred-Page Document Parsing
- Counting the Bandages: On measuring the time that AI saves
- Enhancing Deep Learning Model Performance through Data Generation Strategies: Classification of Olive Tree Diseases
- The Trinity:Capability-Driven Recursive Self-Improvement for Multimodal Agents
- CohortFlow: A Steerable Clinical Cohort Curation Agent with Localized Semantic Correction
- Why and How to Do Science When AI Is Better at It: A Framework Grounded in Human Understanding and Direction
- From Structural Feedback to Prompt Policies: Learning Faithful Text-to-Image Prompt Editors
- Don't Pick a Winner: Knowledge Conflicts Need a Prior, Not a Rule
- Learning Human Sports Motion Feasibility via Quantile Regression on Tracking Data
- Skill Dreaming: Evolving Repository Skills through Pull Request Streams
- MARQ: A Modular, LLM-Agnostic Requirement Quality Assessor Infused with Systems Engineering Knowledge
- BehaveRec: When Does Long-Term Behavioral History Help Sequential Recommendation? A Study with Session, History, and Community Signals
- A Smart Instant Book Filter Model based on Dual-Objective Optimization
- Spiking Neural Networks Are Harder to Steal: A Benchmark Study of Model Stealing Attacks
- When Retrieval Is Not Enough: Obsolete Facts in Agent Memory
- Ontology-guided multi-agent reasoning for explainable differential diagnosis
- How Stable Are LLM-Generated Proofs Under Program Refactoring? A Stress Test for Verifiable Code Generation
- Plenty of Room in the Middle: Revealing Opportunities on a GPU Cluster via Fine-Grained Monitoring
- OA-RAG: Ontology-based Agentic RAG for Faithful LLM Reasoning
- WAKMan: Workload Aware KVCache Manager
- LoRA on a Budget: Function-Aware Post-Hoc Compression for Adapter Fleets
- MAREval: A Multi-Agent Framework for Evaluating Natural Language Recommendation Explanations
- kw-bench: Breaking Frontier Harnesses on Real-World Tasks
- Verifiable Efficiency Limits of Innate LLM Planning in Sliding Puzzles
- LM-DSB: A Lightweight Multi-agent System for Automated Daily Safety Bulletin Generation in Construction
- Your Agent Acts Differently When You're Around: How User Presence Changes Agent Behavior in Negotiation
- RouteDRAG: Learning to Route and Control Agentic Execution for Multimodal Document RAG
- From Solver Trajectories to Teaching Trajectories: Cognition-Aligned Reasoning Distillation
- A Heart Means Recursion: Exploring Visual Meaning and Communication in VLMs
- Correction or Manipulation? Measuring Epistemic Vigilance in LLM Agents Under User Pressure
- When to Intervene? An Empirical Study of Errors and Self-Recovery in LLM Agents
- CogArena: Benchmarking Multimodal Agents on Interactive Behavioral Experiments
- AC-JEPA: Anchored Cross-Modal Joint Embedding Predictive Architecture for Visual-to-Symbolic Translation
- MedConstBench: Benchmarking How Constitutions Steer LLMs in Medical Ethics
- ATEMPO: Amortized Transfer Entropy Model Per-lag Over-Time
- Monitor Accuracy Isn't the Answer: A Decision-Value Framework for Evaluating AI Agent Restart Monitors
- CASEL: Consequence-Aware Selective Execution for Reliable Long-Horizon Tool-Using Agents
- ADTGuard: Adaptive Depth-Time Safeguard for Streaming Inference in Large Language Models
- Follow the Herd: A Data-Driven Approach for Actionable Recourse
- InfraDev: Can Coding Agents Develop AI Infrastructure, and How Can We Tell?
- Learning to Formalize: Bridging Open-Ended Workloads and Fast Decision Models
- God Does Not Play Dice: Enforcing Batch Invariance in Agentic Kernel Generation
- Polarity--Time Interaction and Multi-Level Reliability for Event-Based Spacecraft Pose Estimation
- The uneven rise of AI-generated web content
- Remembering Is Not Enough: Refining Memories and Learning When to Use Them in Multi-Agent Systems
- The View from Nowhere: Population-Centered Agent Evolution
- Learning Where to Trust: Reliability-Guided Expert Routing and Velocity Gating for 3D Deformable Registration
- PsyJev: Psychological Analysis via System-1 Judgment
- Beyond the Evidence Coherence Trap: Agent-Driven Evidence Selection for Temporally Disordered Fake News Detection
- Can LLMs Reliably Grade Olympiad Proofs? A Study of Mathematical Verification with LLMs
- Batching the Reflective Optimization Loop: Parallel Proposals Make GEPA Faster and Better
- Dependency Graphs of Mathematical Arguments: Representation, Structural Measurement, and Proof Review
- An Incremental Knowledge-Graph Learner Digital Twin for Evidence-Grounded Early Warning
- Who Knows What - and What Does It Support? Perspective-Bounded Evidence Calibration for Multimodal Theory of Mind
- When the Critic Falls Behind: Co-Evolving Rubrics and Harnesses for Continual Agent Self-Improvement
- Conservative Policy Iteration For Zero-Sum POSGs
- Evolving Agent Teams
- GUI-JEPA: Candidate-Conditioned JEPA Latents as a Fast Action Policy for Mobile GUI Agents
- RubricAtlas: Mining Type-Level Rubrics from Expert Reports to Evaluate Deep Research Agents
- Noise Steering: Exploiting Wake Steering and Acoustic Directivity for Noise-Constrained Wind Farm Control via Reinforcement Learning
- Natural Hard-Negative Mining from Positives for Instruction-Following Retrieval
- Vision-Language Agents Can Learn in Their Own Latent Worlds
- Do bots feel bad? Assessing empirical evidence for functional valence in LLMs
- Proactive Memory for Event-Driven LLM Agents
- AutoVend: Benchmarking LLM Agents on Long-Horizon Autonomous Vending Operations
- SpectRAG: Weighted Chunk Graphs for Token-Budgeted Context Selection in Multi-Hop Retrieval-Augmented Generation
- Guided-RLVR: Guided Exploration in Reinforcement Learning with Verifiable Rewards
- AI-Assisted Discovery of CNOT Decompositions for Quantum Permutations on Constrained Topologies
- Correctness Conditioning as Selection in Chain-of-Thought Faithfulness Evaluation
- OpenGray: a benchmark for evaluating the decisions of radiotherapy planning agents
- Prospector: learning a compact graph traversal policy for scientific reasoning
- SFE-O: An Ontology for Knowledge-Grounded AI in Supercritical CO$_2$ Biomass Extraction and Fractionation
- WorkFactory: Turning Professional Intents into Executable Work for Scalable Agent Training
- LiveSearch: Questions as Executable Verifier Code for a Live Benchmark of Professional-Source Financial Search Agents
- R\`enyi-Regularized Flow-Based Variational Inference for Bayesian Calibration
- Does Your Generalist Skill Optimizer Really Incentivize Agentic Tasks? Rethinking Skill Optimization via Bipartite Attribution
- GraphRAGProv: GraphRAG Meets How-Provenance
- Grounding LLM Agents in Business Processes: Process-Aware Knowledge Graphs for Reliable Multi-Step Service Invocation
- TCM-DiagCommit: A TCM-SD-Derived Benchmark for Evidence-Conditioned Syndrome Selection and Commitment
- RPSD: Retrieval-Privileged On-Policy Self-Distillation with Optimal Transport
- ReForest: Interpretability Meets Reasoning in Isolation Forest
- Aligning Institutional Power and Operational Ability
- Uncovering Why LLMs Fail: From Internal Mechanisms to Adversarial Reasoning Trajectories
- Agent-Conditioned Routing: Diverse Reasoning Paths Within One Frozen Model
- Clearing the FoG: Training KG Agents to Retrieve the Evidence Behind Their Answers
- Constraining the Commit Boundary: A DIO/ODI-Informed Authorization Guardrail for Agentic Entity Resolution
- Beyond the Readout: A Latent Numerical Frame Shared Across Modality and Operation
- M4Bench: Evaluating Procedural Specification for Clinical EHR Derivation Agents
- Concordat: Typed Sessions for LLM Agents
- From Interaction to Capability: Task Advancement as State and Supervision for GUI Agents
- TeamGym: A Gym for Team Decision-Making by AI Agents with Distributed Information
- Timestral
- Rare, Relevant, and Still Wrong: How Memory Fails LLM Group Decisions
- What Improves During Pretraining? Tracking Knowledge Acquisition with Point-in-Time Models
- KPPax: preprocess differentiable kinetic ODE for machine learning in atmospheric chemistry
- ASTROLABE: A 3D Particle-Level Surrogate for Cosmology Simulation with Physical Flow Matching
- scPyviewer: a Python-native interactive viewer from AnnData single-cell data
- Beyond Greenwashing: Algorithmic Resilience and the Non-Linear Governance Premium in Sovereign Capital Flows
- Task-Utility IQA: Rethinking Image Quality Assessment in Embodied Tasks
- Physiognomy Revisited: How Vision-Language Models Encode Sensitive Attributes Through Image Transformation
- ReCoPII: A Relation-Grounded Framework for Building Contextual PII Masking Datasets
- SCOPE: Source Credibility and Outlet Profiling Engine
- Who Convinced Whom? Tracing How LLMs Change Their Minds in Multi-Agent Systems
- Curriculum-Driven Deep Reinforcement Learning for Autonomous Far-Range Rocket Landing
- ResCom: Activation Residual Compensation for Large Language Model Quantization
- Structured Time-Frequency Representation for Channel Prediction
- AgentBaton: Efficient Model Collaboration in Agentic Tasks
- A Hyper-Logical View on Multi-Agent Path Planning
- EMMA: Structured Episodic Memory for Multi-Agent Workflow Synthesis
- Multiple Leak Localization in Water Distribution Networks through Graph Neural Networks
- UnityBench-Mini3D: Evaluating LLM Agents for 3D Game Development in Unity
- Generation and Verification: Human-Agent Collaboration for Safety-Critical Automotive Software Engineering
- Let the Gate Decide: Per-Lookup Precision for Quantized Conditional Memory in LLMs
- PROBE: Frontier Coding Agents Find Different Bugs Than Maintainers Fix
- Layerwise Repacking: Variable-Length Hybrid-Attention Training without an Extra Redistribution Pass
- Pricing Cross-Bank Context in Federated Anti-Money Laundering
- PolicyMediator: Multi-Regional Policy Learning through Mediation Agents with Evolving Skills in OPC Ecosystems
- MedJev: Evaluating Zero-Shot System-One Decision Making in Medicine
- DuplexTune: Coupled-State Replay for Memory-Efficient Mixture-of-Experts Adaptation
- CIVER: Causal Intervention Verification for Reliable Code Review
- Do Coding Agents Still Need a Debugger? Frontier-Model-Based Fault Localization at the Accuracy Ceiling
- Rollback: Experience Backtracking in World Models for Policy Improvement
- Log-To-Leak: Prompt Injection Attacks on Tool-Using LLM Agents via the Model Context Protocol
- Dirichlet Process Mixtures within Semiparametric Bayesian Networks: A Novel Conditional Density Estimation Approach
- Scaling and Renormalization in Neural Networks: An Effective Renormalization-Group Description of Grokking
- Context Is a Pipeline, Not a Window: Information Attrition in LLM Agent Harnesses
- When Connectivity Misleads: Guiding Agentic Search with Contrastive Evidence Paths
- Physical structure as an organizing principle for universal Hamiltonian learning
- Replication or Corroboration? An Axiomatic Framework for Evidential Dependence in Multi-Agent Debate
- Stratum: Repository Memory for Coding Agents Beyond the Context Window
- Can AI Guide Testing Without Generating Tests? A Comparative Study of AI-Assisted Unit Testing
- Beyond the Estimate: Wiener-Informed Decoding for Easier Training of Pixel-Space Flow Models
- count-catcher: Lean Theorems from Model Counting Certificates
- Base Models Can Detect Hidden Intentions that Prompted Classifiers Miss
- Concorde: Geometry-Aware Link Prediction via Decoupled Energy Minimization
- Counterfactual Profile Perturbations for Recommendation Explanations: A Reproducibility Study
- Structure, Not Sequence: Tracking and Validating Folding Mechanisms in ESMFold's Trunk with Sparse Autoencoders
- Directional Statistics on Circles and Spheres: A New Angle on Data
- Agent or Human? Detecting Commits by AI Coding Agents in the Absence of Authorship Signatures
- Freeze, Don't Clone: Fine-Tuning with Expressive Policies Without Retaining Offline Data
- cuArms: GPU-Accelerated Conflict-Based Planning for Multi-Arm Manipulation
- BUSTA: Teaching Small Language Models to Reason About Binary Execution
- Detecting and Mitigating Query-Based System Prompt Leakage Attacks
- CrowdCC: Jointly Benchmarking Corruption Robustness and Deployment Quantization in Crowd Counting
- Enhancing Clinical Classifier Robustness with Deep Representation Learning Based OOD Detection on DNA Methylation Data
- Reliability-Aware Short-Term Glycemic Risk Prediction Using an Anchor-Based Comparative Framework of Sequence, Multimodal, and State-Updating Digital-Twin Models
- HIDRA: Hierarchical Dual-Routing Attention for Replay-Free Lifelong Imitation Learning
- SOMA: A Benchmark for Fine-Grained Egocentric Behavior and Body-Motion Understanding
- Expert lithological labels do not improve learning when used to predict marine sediment properties
- Repulsive Population Policy Gradients for Behaviorally Diverse Reinforcement Learning
- LociArt: A Unified Paradigm for Free-Style Local Video Stylization
- Full Collapse Is Suboptimal: Controlled Partial Neural Collapse for Long-Tailed Recognition
- FIDES-Live: Evaluating Behavioral Concordance in Sequential Trading Agents
- Proposing Better Rollouts for On-Policy Distillation
- ProCerDCL: Query-Efficient Black-Box Probabilistic Verification of Deep Learning Compiler Optimization
- Do AI Scientists Know What's New?
- RoboAuto: Autonomous Robot Program Search through Multi-Level Physical Understanding
- FrameForge: Harnessing Priors of Video Foundation Models for Unified Image Generation
- REACH: An RL-based Hierarchical Humanoid Crawling Framework for Adaptable Confined Space Traversal
- NextEval: An Agent-Based Solver and Benchmark for Derivative-Free Optimization
- Less is more: learning what to profile for GPU kernel optimization
- MM-Net: A Multimodal Multi-Task Network for Sleep Staging and Sleep Apnea Assessment After Ischemic Stroke
- Benchmarking LLM Agents for Behaviorally Adequate Test Generation from Software Requirements
- Balanced State Subspaces for Mamba-2 State Compression
- Dream-COP: Guiding Algorithm Discovery for Combinatorial Optimization Problems in an Experience World
- Adapting Video Models for Robot Control via Motion-guided Asymmetric Prediction
- Tap2Listen: Instant Target Switching for Streaming Audio-Visual Speaker Extraction
- A Quantitative Approach to Asch Conformity in Frontier Large Language Models
- Look Only When Needed: Certificate-Guided Visual Routing for Efficient Diffusion VLMs
- Atom as an Agent: Emergent Order by Collective Predictive Coding
- Specialization, Diversity, and Scaling in Child-Scale Language Models
- TOPS: Reshaping Flow Source and Transport for Multi-Task Robot Manipulation
- HeteGCD: Statistics-Aware Dual-Prototype Learning for Generalized Category Discovery on Heterophilous Graphs
- Conservative Dueling Bandits: Preference Learning under Baseline Constraints
- From Voiceprints to User Entities: Online Speaker-Memory Construction for Multi-User Spoken Agents
- Generative Refinement Learning for Temporal Interpolation in Scientific Machine Learning
- Robust LLM Data Selection via Content-Invariant Gradient Attribution
- Making Self-Driving Laboratories Auditable:A Unified Framework for Hybrid Actions,Measurements, and Safe Execution
- The Vocabulary Is Already Here: Adapting Pretrained Interpretable Features for Generative Recommendation
- Select-TTT: Query-Conditioned Selective Fast Weight Updates for In-Place Test-Time Training
- An Algebra of Transformers
- SetGRPO: Calibrating Grader-Induced Policy Optimization
- TAPESTRY: A Single Backbone for Video Generation and Robot Control
- When All Rollouts Fail: Policy-Compatible Joint Decoding for Hard-Problem Reasoning
- ActiveAlign: Aligning Egocentric Manipulation via Active Perception
- AdaGround: Token-Normalized Adaptive Zooming with Verifier-in-the-Loop Refinement for GUI Grounding
- Developing a Speech Language Model for Basque
- Beyond QASecClaw: Agent Scaling, Verifier Feedback, and Evidence Reliability in Software Assurance
- Self-Keyed In-Parameter Memorization for External Memory-Free Dialogue Recall in Frozen Language Models
- DCKV: Learned Transform Coding of the Prompt KV Cache in Hybrid Language Models
- Dense Representation Learning and Differentiable Shape Modeling for Spine Landmark Detection
- PailitaoMMSearch2.0 Technical Report: Scaling Generative Retrieval to Unified End-to-End Multimodal Search
- Advancement or Acceleration? How Data Curation Improves LLM Training
- LearnPrunerV2: Learnable Progressive Token Pruning for Vision-Language Models
- Evaluating Priority Dispatch for Premium Membership in On-demand Taxi-sharing
- V-LUMEN: Visual Lookup Memory for Embedding Scaling in Vision-Language Models
- Crop Yield Prediction Using Regression Models: A Comparative Evaluation of Model Performance
- Interpretable Deep Learning Models for Breast Cancer Diagnosis Using Medical Imaging
- Zooming without Zooming v2: From Seeing Details to Connecting Evidence for Generalizable Multimodal Perception
- Blindfold: Keeping the Audit Coin Secret from LLM Agents
- Exploring the Limits of the Delta Learning Hypothesis: Single-Pair Generator Biases in Direct Preference Optimization
- EnergyCal: Taming Loss Spikes in Muon-Family Optimizers for Sparse MoE Training
- Learning to Look Beyond Screens: Transferable Visual Observation from GUI Grounding
- Hardware-Matched Spiking Neural Networks for Real-Time Purkinje Cell Spike Classification
- A First-Step Bias Restores Lost Diversity in Diffusion Model Distillation
- MRI-to-CT Translation of Spinal Images via Anatomy-Guided Pseudo-Pairing and Adversarial Diffusion
- What the Primary Lund Plane Omits: A Model-Stitching Diagnosis of a Jet-Tagger Representation Gap
- Generative Replay for Equilibrium Propagation: Toward Edge-Friendly Continual Learning in Hopfield Networks
- Beyond Parameter Protection: Fisher-Shaped EWC for Replay-Free Continual Learning
- GCSR-OCC: Geometry Calibration and Semantic Routing for Open-Vocabulary Occupancy Prediction
- MLE-World: Predicting Machine Learning Experiment Outcomes with a Learned World Model
- Action-Value Calibration via Continuation Sharing for Long-Horizon Agentic Reinforcement Learning
- Retrieve to Act: Motion Primitive Graph Retrieval for Robust VLA Execution
- Spine: Compiling Subtask Trees into Structured Context for Long-Horizon Coding Agents
- When Is a Relation Worth Trusting? Manipulation Costs, Exposure, and the Trust Trap in Multi-Relational Graph Learning
- Aligning Large Language Models for Safety in African Languages: A Case Study of Hausa
- Where Should the Evaluation Budget Go in LLM-Driven Agent Scaffold Evolution?
- A Lightweight Attention-Based CNN with Saliency-Map Interpretability for Skin Cancer Detection
- AnalogScout: Simulation-Efficient PVT-Aware Analog Sizing with LLM Agents
- WCC: A Learning-Based Framework for Deciding When to Apply Cooperative Coevolution in Large-Scale Optimization
- CoS-VQ: Coordinated Shared-Scale Vector Quantization for Large Language Models
- Channel Selection for Mixture-of-Experts Decoding Needs No Calibration Data
- JEPA-Inspired Latent Field Prediction for Sparse-Pilot MIMO-OFDM Channel Reconstruction
- Learning Quality-Controllable Robot Policies with Iterative Hindsight Relabeling
- Bayes-Adaptive MDPs for Rollout Allocation to Enhance GRPO Training
- FeatureArc: Benchmarking Coding Agents on Evolving Features
- Representing the Unseen: Rethinking Visual Imagination for Spatial Reasoning
- Test-Time Communication in Trading Agent Swarms
- Autonomous fixed-wing interception using reinforcement learning
- PerceptRL: Making Visual Perception Rewardable in MLLMs from Answers to Interactions
- How Reliable Is Individualized Functional-Network Attribution in Resting-State fMRI Cognitive Prediction in ADNI?
- HGPRGDel: GPR-Aware Learning and Ranking for Growth-Coupled Gene Deletion Design
- Follow the Gradient: Tracing Safety Degradation Back to Training Data
- Strategyproof Deep Reinforcement Learning for Online Resource Allocation
- Orthogonal and Parametrized Multi-Attribute Steering of LLMs
- The Structure Gap: When More Data Cannot Replace Structured Reasoning
- Which Hand Was Late? Diagnosis for Robot Failure Recovery in Bimanual Tool Use
- GroundControl: Satellite-Conditioned Multi-View Diffusion for Reconstruction-Free Novel View Synthesis
- When, Where, and What to Correct: Human-Guided Adaptive Residual Correction for Robot Policies
- Kinematic Constraint-Referenced Evaluation for Articulated Object Manipulation
- When uncertainty quantification fails: structural vulnerabilities and corrections for selective prediction under clinical class imbalance
- Gym-Anything-RT: Scaling Real-Time Environments for Vision-Language Agents
- REGER: Modeling inter-item Relations with Semantic IDs for Generative Recommendation
- Understanding the limits of latent world models
- GAS-PFN: Graph-Aware Spatiotemporal Prior-Data Fitted Networks for Zero-Shot Traffic Forecasting
- MoI-Bench: A Benchmark for Image Generation and Editing on Mobile Devices
- Beyond Answer Changes: Matched Same-Class Controls for Local Audio Evidence
- VideoLVR: Latent Visual Reasoning over Grounded Video Evidence
- MuonGS: Pretraining LLMs with Relaxed Weight-Norm Control
- Can LLMs Uncover Hidden Physical Laws through Free Exploration?
- Attention as Lattice Gauge Theory: Relations Interact Only Through Wilson Loops
- ThinkWorld: World Imagination Reinforces Chain-of-Dynamics Reasoning
- Honeycomb-Surfels: Compact Hexagonal Primitives for Efficient Scene Rendering and Reconstruction
- Epsilon-Identifiability of Causal Quantities
- Data-Driven Maintenance Optimization: Aligning Prognostic Learning with Decision Objective
- HiMA-Ecom: A Long-Horizon Memory Benchmark for Hierarchical Multi-Agent E-commerce Assistants
- Re4: Scientific Computing Agent with Rewriting, Resolution, Review and Revision
- Difficulty-Estimated Policy Optimization
- Distilling Metacognitive Behaviors through Agentic Editing
- Optimal Experiments for Partial Causal Effect Identification
- PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement
- Moderation by Interrogation: Uncovering Concealed Intent in Agent Social Networks
- CLEAR: Coupling Local Evidence with Affective Representations for Multimodal Emotion Recognition in Conversation
- From SGD to Muon: Adaptive Optimization via Schatten-p Norms
- Hedge-Bench: Benchmarking Agents on Hard, Realistic Tasks Pertaining to Financial Reasoning
- HOLogicBench: Evaluating Higher-Order Logical Reasoning in LLMs
- Experience Memory Graph: Extracting Reusable Correction Knowledge from Agent Trajectories
- LISA: Linear-Indexed Sparse Attention for Efficient Long-Context Reasoning
- Wavelet Phase Diffusion for Structurally and Semantically Consistent Sim-to-Real Translation
- ScalableRAG: High-Quality RAG with Exact Corpus Operations at Minimal Ingestion Cost
- Tracing the Cascade: A Topology-Aware Evaluation Framework for Scientific Agent Hallucinations
- Hard Constraints, Smooth Gradients: Learning Feasible Inventory Policies via Differentiable Projection
- AI-guided deep visual proteomics reveals recurrence-associated morpho-molecular signatures within triple-negative breast cancers
- NxN E-valuation: Hypothesis Certification via a Conformal CRT Null
- MemPrism: Task-Conditioned Relational Memory Views for Long-Horizon Agents
- Validating FKG.in: Soundness Assessment in LLM-Augmented Indian Food Knowledge
- Benchmark Radar: A Living Database and Search Engine for AI Benchmarks and Evaluation
- When Rubrics Fail: Hallucinations Reveal Blind Spots in Medical AI Evaluation
- Lightning Weave: Improving the Accuracy-Efficiency Frontier of Reasoning Models through Capability Composition
- Tailored to you: longitudinal effects of personalising language models
- Synthetic Hospital: An Open, Verifiable, Physician-Validated Longitudinal EHR Benchmark
- Learning response-aware patient dynamics for respiratory support
- EvoIn: Bridging Evolution and Internalization for Agent Fine-Tuning
- Learn from the Gap: Differential-Aware Advantage Pruning with Adaptive Rollout Sampling for GRPO
- Diversity Combining for Multi-Path LLM Reasoning
- Provenance-Aware Belief Updating in Human-AI Feedback Loops
- Annotrace: A Workbench for Step-Level Annotation of Agent Traces
- Towards Human-AI Complementarity for AI Safety
- Fig: Automated Multiagent Behavior Elicitor
- ARCANA: A Benchmark for Abstraction and Analogical Reasoning
- PaperFormalizer: Author-in-the-Loop Autoformalization of Research Papers in Lean
- AOS-Bench: Can LLM Agents Manage Dynamic Workloads?
- Seeing the Options: Typed Decisions over Visual Candidates with Reusable Option Memory
- Local Entropy Fluctuations for Black-Box LLM-Generated Text Detection
- An actively trained neural operator for near-explicit atmospheric chemistry
- Agentic Algorithm Design: LLMs that Propose, Formalise, and Prove Runtime Bounds for Evolutionary Algorithms in Lean
- Inferring Inference
- DyRA: Portable Dynamic Resolution Adjustment Network for Existing Detectors
- Fairness Definitions in Language Models Explained
- She Works, He Works: A Curious Exploration of Gender Bias in AI-Generated Imagery
- HateSense: Hierarchical Sentiment Analysis Framework for Hate Speech Detection: Implementing Binary and Multiclass Classification Strategy
- The Price of Failure Risk: Clipped Barrier Policy Gradients with Convex Regularization
- Smart Surveillance: Deep Learning and Hybrid Approaches for Dynamic Scene Analysis, Object Detection and Motion Tracking
- Correcting Before Adapting: Gradient Correction for Federated Adaptive Optimization
- AnomalyGFM: Graph Foundation Model for Zero/Few-shot Anomaly Detection
- An Embarrassingly Simple Data Selection Strategy for Million-Scale Multi-Modal Instruction-Tuning
- In-Context Learning for MIMO Turbo Receivers
- On the Reliability and Stability of Selective Methods in Malware Classification Tasks
- Doubly Robust Alignment for Large Language Models
- Role-Augmented Intent-Driven Generative Search Engine Optimization
- How K-12 Teachers Adapt Generative AI to Local Classroom Demands: Cultural Distance Between GenAI Defaults and Everyday Teaching Practices
- From Legacy Fortran to Portable Kokkos: An Autonomous Agentic AI Workflow
- MultiFair: Multimodal Balanced Fairness-Aware Medical Classification with Dual-Level Gradient Modulation
- Semantic World Models
- Cross-User Poisoning: User-Task Boundary Failures in Multi-User Collaborative Language Agents
- SELF: A Robust Singular Value and Eigenvalue Approach for LLM Fingerprinting
- The Theory of Strategic Evolution: Three Principles of Strategic Replicators
- Detailed balance in large language model-driven agents
- Counting Clues: A Lightweight Probabilistic Baseline Can Match an LLM
- SkipCat: Rank-Maximized Low-Rank Compression of Large Language Models via Shared Projection and Block Skipping
- TS-DP: Reinforcement Speculative Decoding For Temporal Adaptive Diffusion Policy Acceleration
- Reach-Avoid Differential game with Reachability Analysis for UAVs: A decomposition approach
- Foreground-Aware Dataset Distillation via Dual-Path Patch Selection
- The Fake Friend Dilemma: Relational Trust and the Political Economy of Conversational AI
- Beyond the Dirac Delta: Mitigating Diversity Collapse in Reinforcement Fine-Tuning for Versatile Image Generation
- SQuID: A Benchmark for Quantitative Spatial Reasoning on Satellite Imagery
- VDE Bench: Evaluating The Capability of Image Editing Models to Modify Visual Documents
- A swap-adversarial framework for domain generalization in pre- and post-stimulation Parkinsonian state classification
- MIDAS: Mosaic Input-Specific Differentiable Architecture Search
- AgentSentry: Mitigating Indirect Prompt Injection in LLM Agents via Temporal Causal Diagnostics and Context Purification
- Flowette: Flow Matching with Graphette Priors for Graph Generation
- Efficient Token-Level Policy Optimization for Masked Diffusion Models via Velocity Alignment
- Preconditioned Flow Matching
- SWE-Skills-Bench: Do Agent Skills Actually Help in Real-World Software Engineering?
- When AI Agrees Too Much: Manipulating Automated Code Review via Bias Injection
- A Computationally Efficient Multi-Section Whole Slide Image Analysis for Capturing Prostate Cancer Multifocality in Biochemical Recurrence Prediction
- Ctrl-A: Control-Driven Machine Learning using Adaptive Online Data Augmentation
- SemLoc: Structured Grounding of Free-Form LLM Reasoning for Fault Localization
- One-Shot Localisation of the Global Minimum of a Noisy One-Dimensional Function: A Controlled Comparison of an Iterative Neural Minimizer, Set Transformers and Classical Estimators
- Clinical Note Bloat Reduction for Efficient LLM Use
- Latent Abstraction for Retrieval-Augmented Generation
- Scoring-Rule Sensitivity in LLM Evaluation: Evidence from Indian Financial Regulatory Text
- AmaraSpatial-10K: A Spatially and Semantically Aligned 3D Dataset for Spatial Computing and Embodied AI
- DeGenTWeb: A First Look at MGT-dominant Websites
- On the Optimal Sample Complexity of Offline Multi-Armed Bandits with KL Regularization
- Posterior-First Neural PDE Simulation: Inferring What Downstream Tasks Need from a Single Field
- Dr. Post-Training: A Data Regularization Perspective on LLM Post-Training
- On the Tradeoffs of On-Device Generative Models in Federated Predictive Maintenance Systems
- RePO-VLA: Recovery-Driven Policy Optimization for Vision-Language-Action Models
- REALM: Retrospective Encoder Alignment for LFP Modeling
- Edge-AI-Driven Learning-to-Rank for Decentralized Task Allocation in Circular Smart Manufacturing
- The Geometry of Learning: Read-Write Asymmetry Across Training
- BoLT: A Benchmark to Democratize Black-box Optimization Research for Expensive LLM Tasks
- LLM Code Smells: A Taxonomy and Detection Approach
- Hidden-State Privacy Has an Empty Middle
- Beyond Questions: Evaluating the Stated Knowledge of LLMs
- Does The Way You Plan Matter? An Empirical Study of Planning Representations for LLM Web Agents
- CalArena: A Large-Scale Post-Hoc Calibration Benchmark
- Grasp-Then-Plan with Failure Attribution: A Closed Two-Stage Framework for Precise and Generalizable Robotic Manipulation
- Streaming Communication in Multi-Agent Reasoning
- DRIFT: A Residual Flow Adapter for Decoding Continuous Outputs in Vision-Language Models
- LLMs Can Leak Training Data But Do They Want To? Evaluating Memorization Propensity in LLMs
- What Makes Video World Model Latents Action-Relevant: Prediction over Reconstruction
- Spatio-Temporal Audio Language Modeling for Dynamic Sound Sources
- DifFRACT: Diffusion Feature Reconstruction and Attribution for Circuit Tracing
- LineageMark: Multi-user White-box Watermarking for Contribution Tracing in Model Derivation Chains
- TokenMinds: Pretrained User Tokens and Embeddings for User Understanding in Large Recommender Systems
- Trust Region Policy Distillation
- From Language to Navigation Goals: A Vision-Language Approach for Semantic Navigation of Mobile Robots Using RGB-D Perception
- From Deterministic to Generative Deep Learning for Urban Air Quality Reconstruction from Sparse Observations
- TextCloak: Safeguarding Data from Unauthorized LLM Fine-Tuning via RL-Driven Unlearnable Examples
- Actionable CBFI: Prescriptive Interventional Diagnosis and Interaction-Aware Escalation for Sparse Causal Algorithmic Recourse
- From Perspective to Fisheye Depth Estimation and Open-Vocabulary Segmentation
- Defense-as-Skill: Evolving Runtime Guard Skill for Skill-Augmented Agents
- Can LLMs Design Video Coding Tools? A Case Study on Planar Mode
- Suan: Rectifying Direct Preference Safety Alignment in Large Language Models
- Legislating World-Model-Based Planning with Legal Reasoning
- MUMINS: Metadata-conditioned Uncertainty-aware Medical Image Next-state Synthesis
- DIP-CAD: A Design-Intent-Preserving Agent Harness for Parametric CAD Modeling
- Learning an Anchored Prompt Space for Continual Adaptation of Large Language Models
- GAUGE: Group-Wise View-Inconsistency Rectification for Feed-Forward 4D Tracking
- Interpretable Synthetic Medical Tabular Data Generation for Clinical Decision Support Using Fuzzy Cognitive Maps
- Corrupted but Correct: Why Vision-Language Models Lie to Themselves Internally
- The Score Is Not the Structure: Brain Alignment and Cross-Lingual Transfer
- Comparing GenAI to Machine Learning and Manual Review for eDiscovery
- Where Small Mistakes Matter: Quantifying State Criticality for Imitation Learning
- Agaram 2.5B: The First Tamil-First Language Model Trained from Scratch with Tool Calling. A Playbook for Under-Represented Languages
- Memorilla: Latent Semantic Memory for LLMs
- First-Order LoRA Training with Phone NPUs
- The Shape of the DPO Hyperparameter Landscape: What to Measure, How It Moves with Scale, and When Not to Transfer It
- From Code Generator to Engineering Companion: A Plant-aware Agent for PLC Programming
- CulturalRiddles: A Multicultural Benchmark of Riddles
- Rethinking Multi-View Fusion for Urban Representation Learning
- Knowing When to Stop: The Evidence-Action Gap in Coding Agents
- A Benchmark for Controllable Speaking-Style Captioning in Audio-Language Models
- Evaluating Visual Language Models on Visual Grounding for Visually Impaired People
- SEEK: Skill Extraction with Evolving Knowledge against Frontier Science Agents
- VLA-HWBench: Where the Time Goes When VLA Models Run on Edge Hardware
- Silent and Costly: Characterizing and Detecting Prompt-Caching Mis-Integration in LLM Agents
- Waypoint: Analytically Modeling GEMM for CPUs
- JANUS: Synthesizing Cross-Layer Defenses Against Web Deserialization Attacks
- The Agent in the Middle: How AI-Mediated Access Reshapes Web Vulnerability Exploitation
- Multi-Fidelity Training Reshapes Uncertainty Decomposition in Atomistic Models
- Quantitative Generalization Bounds for Multi-Layer Transformers: A Few Long Strings can Beat Many Short Ones
- Pretraining Knows What, Simulation Teaches How: Compositional Generalization in a Sim-Trained Robot Agent
- I-HITL: Inalienable Human-in-the-Loop
- KORA: Deterministic-First Execution Control for AI Workloads
- Calibrated for Which Policy? Conformal Uncertainty for Fixed World Models Under Intervention
- Distributed LLM Inference via Communication-Efficient Speculative Decoding
- Exponential Wake-Up Overshoot in Explicit Gaussian Point-Process Filters
- From Slices to Volumes: Comparing 2D, 2.5D, and 3D Residual Networks for Pulmonary Nodule Malignancy-Risk Classification
- Knowledge distillation across teacher architectures for compact diabetic retinopathy grading
- h1: A Taste of the Bitter Lesson in Automated Theorem Proving
- Condition-Enhanced Generative Adversarial Network for Aerodynamic Configuration Design of Commercial Aircraft
- Generalization in Learning: Eyring Formula and Predator-prey Model
- Data-driven Discovery of Lie Symmetries and First Integrals with Latent ODEs
- A Graph-informed Kriging model for sample-efficient Bayesian optimization of wind farm layouts
- Exploratory creative pre-ranking: Hierarchical empirical-Bayes Thompson sampling for real-time creative optimization
- Jaxnasium: Embracing PyTrees For Any-Agent, Cross-Suite Reinforcement Learning
- Instruction Capacity: Behavior and Degradation of Language Models under Controlled Load
- Minimax-Optimal Small-Loss Bounds for Online Linear Optimisation on $\ell_p$ Balls
- Beyond Update Counts: A Control Map for Evaluating Test-Time Model Updates
- SpatialPC: Deep Computation with Sublinear Local Credit Propagation
- Dynamical coarse-graining of neural-network training
- Bayesian Evidence Accumulation for Graph-Based Learning in Dynamic Data-Driven Applications Systems
- Three Worlds for the Price of a Logarithm: Baseline-Protected Best-of-Both-Worlds Combinatorial Semi-Bandits
- What Should an Information Bottleneck Encoder Learn? Separating Representation Learning from Rate Control
- EVA-Flow: Learning Shared Chemistry for Unified Environment-Aware Molecular Conformation Generation
- Bridging Exploration and Loss of Plasticity in On-Policy Reinforcement Learning
- Confidence Is Not a Promise: A Reference Audit of Typed Decision Models
- AHEAD: Adaptive Neural Acceleration for Large-Scale Linear Programming
- From Generation to Scoring: Scaling LLM-Based Semantic Labeling at LinkedIn
- SpecLST: Spectral-Enhanced Long-Short Term Spatio-Temporal Multimodal Modeling for Traffic Forecasting
- When Can a Learned Design Surrogate Be Trusted in a Self-Driving Microfluidics Laboratory? An Offline Closed-Loop Framework Built and Tested on Six Public Tables
- Persistence, Exposure, and Credit Routing in Linear Temporal Predictive Coding
- Property-Based Testing for Robust Softmax and Cross-Entropy
- Simulated petrological labels enable probabilistic mineralogy prediction from whole-rock geochemistry
- Learning atomic interactions for the thermodynamics of liquid mixtures
- High-Fidelity Aerostructural Dataset of Transonic Wings in Cruise Conditions
- Split Federated Unlearning
- CoHARL: Co-Evolving Agent Policies and Executable Harnesses with Reinforcement Learning
- Underdamped Anchored Langevin Dynamics
- PRISM: Pathology Reliability In Scarce-label Medicine. A Tile-level Benchmark of Pathology Foundation Models
- Folding Attention Biases into Keys Makes Compacted KV Caches Run on Standard Kernels
- Holomorphic Cauchy Networks for Efficient Stationary Tail Computation in High-Dimensional Reflected Brownian Motions
- Learning from Points and Patches: A Pre-trained Hybrid Attention Transformer for Time Series Analysis
- Recall or Relearn? A Mixture of In-Context Experts for Recurring Concept Drift
- Efficient Quantization Training with Stochastic Rounding via Structured Subspace Randomness
- Normalized Cuts and Data Clustering via Matrix Fractional Programming
- The Batch Size Can Be One: Test-Time Adaptation from Classifier Geometry
- Deep Bayesian Approaches for Interpretable Nonlinear Structural System Identification
- When Does Perturbation Consistency Add Information Beyond Softmax Confidence? Theory and Evidence for the Decision Consistency Framework
- When Is a Cheap Fairness Edit Safe? Screening Post-Hoc Edits in Two-Sided Matching Markets
- TACO-Graph: Automated Construction of Text-Atributed Graphs
- Log-Derivative Trick for Parametric Markov Chains
- RankFormer: Rethinking Time Series Transformers through Representation Capacity Analysis and Progressive Decoding
- Nonlinear Direct Feedback Alignment for Scalable Backpropagation-Free Training
- Does LLM Pre-Training Typically Occur at the Edge of Stability?
- API-TAIL: Detecting Model Substitution in LLM APIs through Temperature-Aligned Likelihood Tails
- LLM knowledge for improved Out-of-Distribution Tabular Prediction - A Cross-Validated Regularisation Approach
- Tail Cues, Principal Corrections: Plug-and-Play Rectification for Open-Set Test-Time Adaptation
- Hyperparameter and Neural Architecture Search via Evolutionary Optimization for Reinforcement Learning: HPNAS-RL
- HiLo-RD: Cross-Budget Self-Distillation for Masked Diffusion Language Models
- Provenance Signatures for Low Rank Adaptations of Image Diffusion Models
- Gaussian Process Bandits for Constraint Black Box Lever Optimization
- TDLM: Adaptive Temporal Integration for Diffusion Language Models
- Algorithmic Monoculture in Machine Learning-based Site-Specific Fertilizer Recommendation
- Live-OPSD: On-Policy Self-Distillation for Proactive Streaming Large Video Models
- On the VC Dimension of Autoregressive Linear Threshold Functions
- GNN-Guided Network Reduction for Centralized Urban Traffic Routing
- CaLLM: Continual Adaptation of LLMs via Non-Parametric Adapter Routing
- Meta Models: Models that Model Learning
- Compote: Collapse-Resistant Multiple Hypothesis Learning For Sequential Recommendation
- On the Sparsity of Direct Preference Optimization: Weight Disentanglement in the NTK Regime
- Self-Regularized Flow Language Models
- Mechanistic Failure Modes of Weight Pruning in Transformer Language Models
- Variational Optimal Control for Diffusion Guidance
- Physics-Guided Flow Matching: A Generative AI Framework for Global Total Electron Content Forecasting
- SimplexLoRA: Dynamic Rank Updating via Alternating Minimization with Simplex Projection
- Sharpness-Aware Muon: Matrix-Aware Optimization with Momentum-Guided Perturbations
- Your Policy Is Secretly a Harness: The Decision Geometry of Agent Interfaces
- Bayesian Inference of Expensive Simulators: Survey and Comparison of Design Choices
- Structured Representation Probing for Risk Anticipation in End-to-End Autonomous Driving
- Normative Deviation Modeling for EEG-based Depression Detection
- EmbeddingGemma 2: Unified and Efficient Multimodal Representations
- TRIDENT-Seq: Profile-Matched Prefix-Future Contrast for Label-Efficient Payment Risk Modeling
- Euclidean Crowding-Driven Adaptive Euclidean-Hyperbolic Graph Learning
- Para-Gaussian Approximations of Random Deep Neural Networks
- How Many and How Much? Contrastive Learning with Negative Samples for Knowledge Graph Embedding
- LinMix: Data Mixing as Linear Domain Attribution
- Displacement Moments and Path-Terminal Separation in Brownian Interpolant Sampling
- When Visual Evidence Is Insufficient: How Instructions, Post-Training, and Internal Activations Shape Refusal in Multimodal Language Models
- CRISP: Co-designing Pruning and Scheduling for Efficient MoE Inference on Edge Servers
- Interpretable Meta-Policies for Neural Decision Making
- An Input Skip Can Reparameterize a VAE Without Improving the Generator
- GTFormer: A Syntax-Free Graph Network Alternative to Transformers for 8192$\times$ Length Extrapolation in Linear Time
- Collective Dynamics of Self-Reinforcement: How Cross-Task Coupling Drives Policy Collapse in Unsupervised RLVR
- Pneuma: A Large-Scale Real-World Time Series Benchmark for Commercial Aircraft Predictive Maintenance
- LCM-Promo: An LLM-Based Uplift Model for Promotion Allocation
- EnvSched: Environment-Aware Scheduling for Rollout-Efficient Agent Reinforcement Learning
- Scaling Laws for Tool-Use Agents via Controllable Environment Synthesis
- Heads of Security: Learned Attention Gates Improve Prompt Injection Robustness
- When Should We Intervene at Test Time?
- Learning to Exploit Passive Dynamics for Energy-Efficient Target Hopping of a Spring-Legged Quadcopter
- Reconstructing the Human Development Index at the Individual Level from Survey Microdata: Development and Validation of a Methodology
- Geometric priors for single-patch neural Calabi--Yau metrics on the quintic: a conservative fixed-$\psi$ Ablation study
- Anchored Stein Variational Gradient Descent Algorithm for Non-smooth Sampling
- Causal DAG-Guided Reinforcement Learning for Long-Term Physical Activity Maintenance: Using Prior Data to Inform Future Deployment
- The proton structure one mode at a time: a Neural Tangent Kernel analysis of a DIS fit
- Learning on-the-fly quasi-stationarity diagnostics from low-dimensional surrogates
- Closed-Loop Aware Planning via Imitation-Free Speed Scoring for End-to-End Autonomous Driving
- FinLLM-KG: Knowledge Graph-Enhanced LLMs for Financial Sentiment Analysis in Algorithmic Trading
- SuperJEPA: Pretraining with a Supervision-Augmented Joint-Embedding Predictive Architecture for High Energy Physics
- Symmetry Breaking Is the Price of Finite Capacity
- Anytime Valid Confidence Bands for Distribution Functions
- LLM Hallucination Detection via Low-rank Multilinear Decomposition
- Approximation Families and the Statistical Efficiency of Denoising Score Matching
- Cross-modal Steering Vectors Exist Across Both Image and Text Generation
- Privately Clipping Heavy-Tailed Data
- FASST: A Compact Stateless Atmosphere Model for Long-Term Ocean Coupling
- Balanced Active Subspaces for Optimal Control via Trajectory Optimization
- HELICS: Biobank-scale Conditional Synthetic Genome Generation via Latent Flow Matching
- IOSage: Detecting HPC I/O Bottleneck Patterns to Guide LLM Recommendations
- Evaluating Federated Vision-Language and Vision-Only Models for Privacy-Aware Breast Cancer Imaging
- ORBITAL: Optimization of Robust Brain--Computer Inference Through Architecture-Aware Low-Precision Compression
- Whole-Body Humanoid Control via Stochastic Motion Tokenization
- A Simple Presentation of Robust Membership Inference Attacks
- Second-order Sinkhorn Geometry and $\sqrt{n}$-Inference
- Algorithmics for Graphical Separation Criteria
- Causal Localization of Evidence Effects in Transformers
- Impossibility of No-Regret in Multi-Agent Bidding
- From Biological Curves to Soft Modular Robots: Learning-Based Inverse Design with Uncertainty-Guided Refinement
- Quantifying and Prioritizing Stealthy Source-Manipulation Threats Against Multi-Source Anomaly Detectors
- SMEBENCH: Benchmarking Neural Music Editing with Synthetized MIDI Edits
- Neuronal Identity as an Organizational Basis for Analyzing Neural Population Dynamics
- Frequency-Uniform Log-Regret Reinforcement Learning for Diffusions
- Variable-Length Feedback Coding With Universal Decoding and Bandit Channel Selection
- Joint Distributional Copula Regression for Insurance Claim Frequency and Severity via Boosting
- Boundary Geometry Determines Optimal Shallow ReLU Approximation under Weighted Coefficient Constraints
- Evaluating AlphaEarth Embeddings for Crop Classification: Separability, Temporal Dynamics, and Cross-Regional Transferability
- HorizON: A Multi-Horizon Deep Operator Network for Property Prediction of Organic Battery Stacks
- Intrinsic Information-Theoretic Analysis of ReLU Nets
- Sample-and-Score: Surrogate-Guided Coordinated Spatial Reuse in IEEE 802.11 Networks
- RAG-CSI: Global Local Retrieval via Joint State and Dynamics Matching for CSI Prediction
- PREEN: Exemplar-Based Color Style Transfer with Smartphone Editor Presets
- Quantum Evolution Kernel for Node-Attributed Graphs
- Online gas classification at the edge
- From Ethernet to Wi-Fi: Flow Feature Selection and Construction for Traffic Classification
- Quantized Edge Inference of a Self-Supervised SAR-Optical Encoder on Commodity x86 Hardware
- Systematic Morphometric Brain Autoencoding for Representation Learning in Neurodegenerative Diseases
- How Much Can Best-of-N Help a World Action Model? Measuring Selectable Headroom and Realizing It in Closed Loop
- Retrospective macromolecular pool fraction mapping from conventional clinical MRI: a physics-guided self-supervised deep learning framework
- Doubly Robust Deep Kernel Tests for Structured Outcomes
- Watermark Basins: Black-Box Detection of Distilled LLMs
- Low-degree polynomials lower bounds for clustering in small dimension
- Scale-aware and spatially-localised optimisation of kilometre-scale MLWP models
- Online Interval Selection with Binary Predictions: Optimal Consistency and a Continuous Robustness Tradeoff
- The Geometry of Manifold Learning: Singular foliations via Deep Learning models
- PDRVLA: A Vision-Language-Action Model with Physical Dynamics Representations
- An end-to-end framework for uncertainty-aware validation of generative models
- Multi-Agent Reinforcement Learning of Co-Adaptive Hip Exoskeleton Control (MARL-Exo) for Metabolic Energy Reduction
- Graph Representation Learning via Adaptive Expectile Aggregation
- QD-PINN:A Physics-Informed Neural Network Approach to Bayesian Quickest Detection in High-Dimensional Diffusions
- Keeping Inference in the Radio: Design and Measurement of a Spiking Accelerator for mmWave ISAC Activity Recognition
- Regression Functions as Fields on Covariate Space: Evolution Equations, the Hessian, and Identification
- A note on data-efficient finetuning of vision foundation models
- Spectral Braids: A Topological Descriptor for Dynamic Directed Networks
- Zero-Shot Multimodal Event Spotting in Soccer Broadcasts
- Pushing Dexterity and Safety for In-Hand Object Pose Reaching That Works Remarkably Well
- Adaptive Momentum Benefits Score-based Generative Models
- A Comparative Study of Convolutional and Recurrent Neural Networks for Storm Surge Prediction in Tampa Bay
- Temporal difference learning adapts to incomplete coverage: an oracle inequality over subgraphs
- Anomaly Detection via Autoencoder Composite Features and NCE
- Targeted Learning for Inference on Data Fairness
- Concise One-Layer Transformers Can Perform Dictionary Lookup
- Exact Unlearning of Finetuning Data via Model Merging at Scale
- Manifold meta-learning for reduced-complexity neural system identification
- StablePCA: Distributionally Robust Learning of Shared Representations from Multi-Source Data
- Data-Driven Laplacian-Constrained Personalized Federated Learning
- FedHQ: Hybrid Runtime Quantization for Federated Learning
- Accuracy vs. Accuracy: Computational Tradeoffs Between Classification Rates and Utility
- Differential Privacy of Decentralized Learning: Centralized Rate Conditions and Nearly Tight Guarantees
- Improved Regret Bounds for Gaussian Process Upper Confidence Bound in Bayesian Optimization
- Development of Deep Learning Optimizers: Approaches, Concepts, and Update Rules
- On The Variability of Concept Activation Vectors
- Orlicz-Sobolev with Musielak: An Efficient Regularization Approach for Graph-based IPM
- Robust GNN Watermarking via Implicit Perception of Topological Invariants
- SeeNav-Agent: Enhancing Vision-Language Navigation with Visual Prompt and Step-Level Policy Optimization
- Efficient Generative Transformer Operators For Million-Point PDEs
- In-Context Multi-Operator Learning with DeepOSets
- Is BatchEnsemble a Single Model? On Calibration and Diversity of Efficient Ensembles
- MapPFN: Learning Causal Perturbation Maps in Context
- Sparse Attention as Compact Kernel Regression
- SplineFlow: Flow Matching for Dynamical Systems with B-Spline Interpolants
- COMPOT: Calibration-Optimized Matrix Procrustes Orthogonalization for Transformers Compression
- NN-OpInf: an operator inference approach using structure-preserving composable neural networks
- Scaling Reasoning Efficiently via Relaxed On-Policy Distillation
- Characterizing WebGPU Dispatch Overhead for LLM Inference Across Four GPU Vendors, Three Backends, and Three Browsers
- Conditional Sampling via Wasserstein Autoencoders and Triangular Transport
- Integrated electro-optic attention nonlinearities for transformers
- Structure-Aware Variational Learning of a Class of Generalized Diffusions
- Geometric Characterisation and Structured Trajectory Surrogates for Clinical Dataset Condensation
- ZeRO-Prefill: Zero Redundancy Overheads in MoE Prefill Serving
- Rethinking State Tracking in Recurrent Models Through Error Control Dynamics
- The Global Empirical NTK: Self-Referential Bias and Dimensionality of Gradient Descent Learning
- Twincher: Towards Robust and Precise Physical AI via Bijective Representation Learning
- SCI-Defense: Detecting Persuasion-Structured Manipulation of LLM-Based Ranking in Generative Engines
- Strong Teacher Not Needed? On Distillation in LLM Pretraining
- Bastion: Budget-Aware Speculative Decoding with Tree-structured Block Diffusion Drafting
- Re-evaluating Confidence Remasking in Masked Diffusion Language Models
- PhysGuard: Fisher-Guided Gradient Projection for Sim-to-Real Neural PDE Surrogates
- Formalizing and Estimating Task-Space Complexity for Zero-shot Generalization
- LiteTopK: Exploiting the Curse of Dimensionality for a Fused Indexer-TopK Kernel in Long-Context Sparse Attention
- Beyond Single-Episode Optimization: Sliding-Window Aware Generative Auto-Bidding for Long-Term Advertising Effectiveness
- Graph-structured joint-embedding prediction for single-cell transcriptomics
- SoftMCC: An MCC-Brier Calibration Bridge for Threshold-Free Model Selection under Class Imbalance
- ProxyGuard: Direct Reliability Inference for Randomized Data Release Mechanisms with Shared Targets
- K\"ahler landscapes for complex neural network descents and guarantees including a search and destroy of the Calabi-Yau manifold
- SPARCL: Spectral Partitioned Analytic Continual Learning
- Revenge of Monosemanticity: Neuron Specialization as a New Form of Feature Learning in MLPs
- MEL: Coordinate-Preserving EEG Tokenization for fMRI Translation
- Structure and Regularization of Fused Objectives in Kernel Discriminant Analysis for Small-Sample-Size Problems
- Beyond Embedding Transfer: Component Roles in Grokking Transfer and Stability
- QuantForge: Discovering Residual Decompositions for MXFP4 Post-Training Quantization
- Non-Commutative Lie-Algebraic State-Space Models: Finite-Division Variety Laws, Semidirect Cascades, and Gradient Saddle Escape
- GARV: Graph-Aware Refinement and Safety Verification for High-Level Robot Plans
- Strategic Multi-Agent Learning for Interpretable Action Valuation of All Players in Football
- Learning Pareto Stationary Fronts via Single-Pass Backpropagation
- Lost in the bf16 Cast: Exporting Ternary Language Models Can Revert Most Low-Learning-Rate Code Changes
- Trajectory-Aware Policy Optimization: Closing the State Distribution Gap in Long-Horizon LLM RL
- Attention-Graph Curvature for the Verification of Mathematical Reasoning
- MemWM: Associative Memory-Enhanced Latent World Models for Long-Horizon Planning
- What Remains Without Leaking Test Labels? Exposing Evaluation Flaws in Graph-Level Anomaly Detection
- Maestro: Agentic Formal Proof Development by Coordination
- Hilbert-Scale Stability of Posterior Diffusion Priors for Ill-Posed Signal Recovery
- BRIDGE: An Acceptance Gate to Block False Improvements in Model Updates, with a Retrospective Clinical Evaluation
- Loopholing All the Way: Language Modeling with pure Self-Conditioning
- Exact Spectral Uncertainty Decomposition for Linear-Gaussian Bayesian M/EEG Inverse Problems
- Benchmarking Model Compression Techniques Across Heterogeneous Edge Platforms: A Predictive Maintenance Use Case for Rotating Machinery
- Same Luck, Different Choice: Twin-World Credit Assignment for Agentic Reinforcement Learning
- A Fix for Softmax and Cross-Entropy
- CuraLoc: Repository-Level Fault Localization via Planned Code Inspection
- Tree Depth and Sample-Quality-Associated Features: Two Hidden Factors in Boruta Selection for RNA-Seq Data Classification
- The Cost of an Unknown Horizon for the Last Iterate of Clipped SGD under Heavy-Tailed Noise
- Climate downscaling with likelihood functional flow matching
- INTRA: Interleaved Non-contiguous Token spaRse Attention
- STeP-TS: Shared Temporal Filters with Pruning for Time Series
- Shared Context, Separate Engines: Streaming Vision--Language Inference on Edge SoCs
- CoPE-CGL: Community-aware Prototype Extraction for Replay-Free Continual Graph Learning
- Auto-Research Arena: Benchmarking and Advancing Auto-Research in Language Modeling
- NAG-GS: A Semi-Implicit Stochastic Optimizer with Large-Step Stability
- Nash Incentive-compatible Online Mechanism Learning via Weakly Differentially Private Online Learning
- A Deep Learning Framework for TTV Characterization: A Case Study of the Kepler-88 System
- Non-asymptotic Estimation of Sparse Partial Correlation via Joint Partial Regression
- A Mechanistic Interpretation of Prescribed-Parameter Convolutional Kernels
- Foreclassing: From Forecasts to Decisions
- Investigating Active Sampling for Hardness Classification with Vision-Based Tactile Sensors
- FRAME-C: A knowledge-augmented deep learning pipeline for classifying multi-electrode array electrophysiological signals
- When is Fine-Tuning Scalable? Certified Recovery from Noisy Data and Linear-Cost Fine-Tuning on Structured Task Shifts
- Detection of trade in products derived from threatened species using machine learning and a smartphone application
- CLEAR-IR: Clarity-Enhanced Active Reconstruction of Infrared Imagery
- Toward Uncertainty-Aware and Generalizable Neural Decoding for Quantum LDPC Codes
- scTREAT: Latent Diffusion for Single-Cell Drug-Response Prediction
- Optimizing Spectral Signature in Code Backdoor Detection
- LIME: Link-based User-item Interaction Modeling with Decoupled XOR Attention for Efficient Test Time Scaling
- Uncertainty-Guided Multimodal Consensus for Multimodal Sentiment Analysis
- Seeing is believing: The case for hardware-centric inefficiency detection and remediation
- Latent Nonlinear Denoising Score Matching for Enhanced Learning of Structured Distributions
- R2CC: Reliable and Resilient Collective Communication for Distributed Learning
- Semantic-Geometric Task Representations for Bimanual Manipulation from Human Demonstrations to Robot Action Planning
- OneFlowSBI: One Model, Many Queries for Simulation-Based Inference
- Learning Sequential Decisions from Multiple Sources via Group-Robust Markov Decision Processes
- STRAND: Sequence-Conditioned Transport for Single-Cell Perturbations
- Learning to Control: The iUzawa-Net for Nonsmooth Optimal Control of Linear PDEs
- Ensemble-size-dependence of deep-learning post-processing methods that minimize an (un)fair score: motivating examples and a proof-of-concept solution
- HiAP: A Multi-Granular Stochastic Auto-Pruning Framework for Vision Transformers
- Machine learning cannot decarbonise the subsurface until it can be audited
- Expert-Guided Class-Conditional Goodness-of-Fit Scores for Interpretable Classification with Informative Missingness
- Discovering quantum phenomena with Interpretable Machine Learning
- Locally invariant manifolds and foliations
- Compliance Moral Hazard and the Backfiring Mandate
- PGID: Progressive Guided Inversion and Denoising for Robust Watermark Detection
- Phases of Muon: When Muon Eclipses SignSGD
- Multi-Variable Conformal Prediction: Optimizing Prediction Sets using Multiple Calibration Variables
- Parametric Rates for Nonparametric Conditional-Moment Bilevel Optimization
- An Asymptotic Theory of Chain-of-Thought in In-Context Learning
- Distributionally Faithful Imputation via Positive Semi-Definite Kernel Density Estimation
- Computational and Statistical Guarantees of the c-Rectified flow
- Parallel Universes for Temporal Evaluation and Knowledge Updating in LLMs
- Momentum in large-batch training: Polyak enlarges the critical batch size, Nesterov improves data efficiency
- A Tight 3-Approximation for Fair $k$-Means
- Self-Supervised Learning for Robust Resonance Mass Regression in Cascade Decays
- Similarity Pairing with Energy Mover's Distance for Self-Supervised Pre-Training at the LHC
- When Does Test-Time Physical Diagnosis Pay?
- Nuclear Norm-Regularized Bayesian Matrix Completion
- Plan-to-Synthesis: Cross-City Human Mobility Generation via Semantic Latent Flow Matching
- CraftSPH: A composable framework for differentiable smoothed particle hydrodynamics implemented in PyTorch
- Advantage of Sample Complexity in Quantum PAC Learning Requires Inverse Access to State-Preparation Unitaries
- SD-DPC: Sparse Dictionary Differentiable Predictive Control
- Denoising Hierarchical Representations: Joint Continuous Diffusion for Language Modeling
- Calibrating LLM Explanations to Human Decisions with Reinforcement Learning
- When Does the Dependency Graph Help Code-to-Architecture Mapping?
- BiDex-100: A Comprehensive Benchmark for Bimanual Dexterous Manipulation Across 100+ Simulation Tasks
- What Prediction-Market Contracts Reveal at Launch: Trading Volume and Dispute Risk
- EgoHand: Monocular 3D Hand Reconstruction from Egocentric Video
- How Much Do You Trust A Dog? Quantifying Hint-Following in Reasoning Language Models
- Combinatorial Lock for Trojan-Resilient AI Engines via Hash-Assisted Spectral Filtering
- Attack logics, not outputs: Concept-based Property Attacks for Shortcut Elimination in Neural Networks
- mdRNAcat: A Molecular Dynamics RNA Catalog of REST2 Conformational Ensembles
- Continue or Restart? The Limits of Early Failure Prediction in Coding Agents
- Testing Agentic Service Systems: An Agentic Living Lab Methodology Across the Physical-Digital Continuum
- Small agents should know what their tools do
- Before the First Edit? Measuring When Coding-Agent Guidance Actually Arrives
- AGDecode: Semantic Attribute Grammar Derivatives for Safety-Constrained Decoding of Large Language Models
- Instant Software Factories
- Your Wish Is My Commit: Over-Compliant Refactoring in AI Coding Agents
- When AI Can Build It, What Should We Assess? Rethinking Evidence of Engineering Competence
- From Ripple to Boundary: Deterministic Impact Boundaries for LLM-Based Software Evolution
- A Fine-Grained Empirical Study of Upstream Security Events and Downstream Mitigation in Maven
- ASNER: Software Architecture Recovery with Adaptive Semantic Neighborhoods and Evidence-Grounded Refinement
- A lightweight C Code Generator for Heterogeneous TinyML Systems
- Does More Thinking Pay Off? Reasoning Cost and Correctness of LLMs in Code Generation and Bug Fixing
- Toward Circuit-Aware Mutation Testing for Quantum Programs: Exact Visibility and Execution Planning
- BenchMeUp: Building Software Engineering Benchmarks from Natural-Language Specifications
- Hidden in Plain Sight: an Empirical Study about The Instruction Stacks of Coding Agents
- Mind the Intent Gap: A Methodology for Collaborative Agent-Driven Software Engineering
- HIDE: Hierarchical Isolation via Documentation and Evidence for Rustc Bugs
- An EVM-Cosmos Bridge for Atomic, Semantic-Preserving Transfer of Environmental Tokens
- PASS Is a Claim, Not an Observation: Mechanical Re-verification of 10 Self-Asserted Fix Claims in a Live Multi-Agent Audit Layer
- Building Mathematical Software -- Lessons from the Development of GAP
- Agentic Refactoring: An Empirical Study of AI Coding Agents
- Bit-Precise Conformance Testing of Simulink Model Checkers
- Structuring license permissiveness from pairwise comparisons
- AI Writes Faster Than Humans Can Review: A Longitudinal Study of an Enterprise 2x Mandate
- Choosing Coding-Agent Benchmarks for Practice: Lessons from Task Demands and Agent Behaviour
- Vibenix: Agentic Software Packaging and Maintenance with Nix
- Beyond Correctness: How Novice Software Engineering Students Exercise Oversight of Generative AI
- Honesty Testing of LLMs and AI Agents: A Vision for Software Engineering
- QTUNER: A Feature-Guided Adaptive Framework for Quantum-Compiler Optimization Pass Tuning
- This is not sustainable
- A fast and secure tar parser for the Rust ecosystem
- Code-execution flaw in UEFI firmware affects most servers on planet Earth
- GOTO Considered Totally Sick (for error handling in C)
- Ghostty 1.4.0
- Four Months of Zig
- Start Doing Graph Theory
- Hoot 0.10.0 released
- Billion-query-per-second lookup tables in C++
- The Benchmarking Crisis
- Forging a Microsoft Edge Full Chain with Logic Bugs Only
- Scalable Vector Grief
- Reminder: Lobsters NYC Meetup Tomorrow 5:30 at Walker's
- 1,447 emoji? We'll do it live
- Raspberry Pi Zero 3 W and Pi 4 Model A on the Way?
- Python's experimental JIT to be removed
- Who’s 0xabadbabe and why? (2011)
- OpenBSD Jumpstart 8.0: Quick Start Guide for New Users
- Pair-coding Lobste.rs feat
- The October inflection point for vibe game development
- Shitty Screenshare
- Making SIMD Code Scale to Any Vector Width
- Compile-Time Dependent Types
- Day 28: Human-in-the-Loop Architecture — Machine Second, Human First in Practice (いいね相当スコア: 0)
- Wrapping Claude Code and Codex in Electron: A Stop Bug (いいね相当スコア: 0)
- Top 21 Sites to Buy GitHub Accounts- Forums (いいね相当スコア: 0)
- Prevent Git merge conflicts by claiming files before you edit them (いいね相当スコア: 0)
- The AI is lying to you. (いいね相当スコア: 0)
- Introducing Sonar (いいね相当スコア: 1)
- Why Local Search Research Should Group Needs by Service, Not Exact Phrases (いいね相当スコア: 0)
- Tailscale SSH Port Forwarding Carries Root Access Across a Tailnet (いいね相当スコア: 0)
- The Gig Economy Reclassification Battle (いいね相当スコア: 0)
- How artificial intelligence fared when it took over a team in English football (いいね相当スコア: 0)
- Agentic Engineering: Be Unreasonable to Your Agent (いいね相当スコア: 1)
- How to Run LFM3-VL-2B Locally: Better OCR Than LFM2-VL in the Same 1.25 GB (いいね相当スコア: 0)
- How to Reconcile LLM API Usage and Allocate Costs by Department: A Reproducible Method (いいね相当スコア: 0)
- Architecting Production-Grade Agentic AI Systems: Memory, Orchestration, and Latency Optimization (いいね相当スコア: 0)
- How to Benchmark LLM API Performance: TTFT, Streaming Stability, and a Reproducible Method (いいね相当スコア: 0)
- How to Verify Your LLM API Provider Isn't Swapping Models: A Reproducible Fidelity Test (いいね相当スコア: 0)
- Your coding agent needs more than instructions (いいね相当スコア: 0)
- How One New RL Framework Slashed AI Agent Latency & Boosted Success by 22% (いいね相当スコア: 0)
- Claude Fable 5.1 API: 1M Context, $0.25/M Cache Reads and Migrating from Fable 5 (いいね相当スコア: 0)
- Claude Opus 5.5 vs GPT-6 Sol vs GPT-6 Luna: API List Prices and Cost per Request (いいね相当スコア: 0)
- El modelo se alquila, el andamiaje es tuyo: cómo hacer un agente de IA que no dependa del modelo (いいね相当スコア: 0)
- My LLM pipeline cited real URLs for numbers those pages never said (いいね相当スコア: 1)
- Discussion Hub for new Claude incident: Elevated errors affecting the Google Calendar connector on Oct 28, 2026
- Is Claude Pro actually worth it, or is the free plan enough? (As a student on a tight budget)
- It happened, my project manager just made a PR with Claude.
- I got Claude Opus 5.5 to render photoreal Blender scenes — it took 3 failed attempts per scene, here’s what actually fixed them
- Claude Code usage exploded 8x today. Anthropic confirmed service errors consume our paid limits — and they won't compensate users
- I think in pictures, so I taught Claude Code to draw mermaid in the terminal
- Claude is driving me nuts
- First Session After Weekly Reset
- PSA for vibe coders: stop prompting tweaks, make Claude build you a remote control
- Claude Opus 5.5 is insane in creating creating 3D movies.
- Not Enjoying Work That Much Anymore
- Anybody else see this bug? Claude not responding to me
- Do you think Claude Fable 5.5 will be released before or after the IPO? Because it is a point of attention for the company that plans an introduction in early November!
- Fast-paced soccer game with solid physics
- So which 5.5 model on which effort should I be using for everyday coding tasks?
- Finally done with all four Claude certs. Three things I wish I'd known going in
- Tried Claude + Blender for the first time today ❤️
- I gave Claude Code ~600 tools to drive Altium Designer. Here it designs a full 4-layer PCB from a written spec
- The lack of transparency from Anthropic is Borderline concerning
- Claude is performing 3.8x poorer compared to last week based on my claude code token analysis
- Claude Haiku 5.5 only?
- Codex quietly added restrictions that took me days to find
- Is anyone actually getting good results with ChatGPT in Roblox Studio?
- How is it possible that GPT-6.1 SOL performs so much worse than Opus 4.7?
- AI Slop Doesn't Prove Anything
- Burned 70% of my weekly Codex limit migrating one website from Vercel to a VPS. Is that normal?
- Codex account switching vs upgrading to Pro — what's more efficient?
- I built an open-source app that hands a coding task to another agent when one hits its usage limit
- Session-start memory injection made my coding agent trust memory over the code
- Need help please! Model Usage Questions
- using Opus or GPT-6 to rename variables and write test scaffolds is peak AI hype waste... (let me explain)
- When I stopped asking AI to do every click
- wall collision issue
- Regarding changing your TV using AI
- My reviewers keep asking why and the honest answer is Codex decided
- With just 1% left to finish this update 😭
- AMA Announcement: MOSS, the Team Behind MOSS-Image, MOSS-TTS, and More (Friday, Oct 30, 8–11 AM PT)
- Mac Studio M5 Ultra 512gb Starts at $22,299.00 USD
- IT vendor lists massive $60 million AI supercomputer on eBay with 9,600 Nvidia A100 GPUs and 4,200 AMD EPYC CPUs, plus free shipping — 300 racks of decommissioned GPUs on offer
- Someone built a 32-bit RISC-V CPU directly inside Qwen's model weights so it can execute instructions internaly!
- Dirt cheap GPUs coming next year
- Gemma 5 models soon ??
- List of suggestions for Gemma 5.
- A small step for man, a GIANT step for LOCAL AI
- nvidia/NVIDIA-Nemotron-3.5-Transcribe-30B-A4B · Hugging Face
- A 4B model made me a netherite ingot from bare hands. All I did was type one sentence in chat.
- AA is getting ridiculous now. Zero open weights in the Most Attractive Quadrant.
- Used Qwen 3.8 flash next on underspeced PC to Root and gut LG TV of spy and adware
- Alternatives to Artificial Analysis
- Working on a way for local LLMs to share context without centralizing everything
- Alternative to artificialanalysis for assessing latest models?
- Theory: Local is going to explode at enterprises
- Diffuspace/Clover-35B-A3B-Preview · Hugging Face
- CUDA: fuse MoE gate/up matmuls and GLU in MMQ by anujj · Pull Request #31188 · ggml-org/llama.cpp
- I tested Strata's routing optimizations with Qwen 3.8FN IQ3_S in coding, math, and instruction following
- I am a King, but local 😂
- I fine-tuned an LLM to clean up transcriptions for my voice-to-text app
- I trained a 21B MoE (3.36B active) from scratch on 70B tokens (emergent ability observations)
- Weekly Thread: Project Display
- I don't understand the usefulness of "Skills"
- Grok Bot, Muse, Dots, Gemini, Duet: five big-tech agents in 10 weeks. An honest founder’s take on what they got right, and why I think none of them will be the last one you set up
- is n8n still relevant?
- Someone please help (retell, cal.com)
- I tested how detectable 23 browser agents are from the website's side. Only one signs its requests
- Follow-up: AI agents keep obeying the hidden note on my sites. Not one of them will say who it is.
- When has an agent asked a question that actually saved you time?
- How to check whether your AI agent's security guard actually blocks what it flags
- My AI can follow its instructions—but often fails to retrieve them when needed. How do you solve this?
- What’s the most useful AI agent you’ve built or set up for your work?
- we have launched a claude agent as process manager with name, slack, email address for processes- shall she become smarter and smarter or would you do one agent after another?
- Agents that pay for their own compute and APIs: smart or a disaster waiting to happen?
- Security asked me two questions: Which MCP Servers are safe for enterprise use, and how do we govern them?
- Week 5 of letting an agent run growth for my side project: someone here told me to stop aiming at the edge of a ban, and the numbers backed them up
- Licensed Merch seller looking for AI Models to make Social Media Content
- What's the best way to connect Claude or ChatGPT Enterprise to our company data securely? Trying to avoid one big service account.
- Any slide generators that output normal powerpoint files?
- How do you stop a browser agent from acting on a page that already changed? We ended up putting the check inside the engine.
- What is something that you have successfully automated with an AI Agent?
- 8 open-source GitHub repos to run a lean AI agent + automation stack in 2026 without stacking $20/month SaaS tools
- How many must-follow rules never reach your coding agent? Do code graphs help? Here's how we measure it.
- How to reduce duplicate actions when your AI agent retries a tool call
- Berkeley paper: agents that dropped their own history beat parallel agents by up to 15%. What do you keep when you reset context
- Researchers found an "evil neuron" and when they activate it... holy shit
- Kinda fucked up
- When will OpenAI change its name to OpenSI?
- How big a deal is this? Claude Opus 5.5: "If these hold up, this is the biggest single day in the history of math and theoretical computer science."
- OpenAI brings its always-on Dots agent into text messages
- OpenAI's GPT-6.1 Astra, Sol, and Luna models are coming to Google Cloud and Antigravity
- Your company wants your personal ChatGPT merged into its workspace. Export everything first, because afterwards there's no export button and no undo
- It seems GPT-6.1 Terra has its place
- Crazy realistic now. We're cooked.
- An agent swarm formalized the 15,000-page classification of finite simple groups in 24 days — the bottleneck was the checker, not the models
- the AI that worked in call centres didn't replace anyone
- Real world comparison of Terra 6.1 vs Sol 6.1 vs Sonnet 5.5 vs Luna 6
- Two physicists asked AI to choose a problem on its own, solve it and write a paper that could be published in a prestigious journal. In two hours, AI produced a paper with a novel, significant result.
- Use GPT on your Hasselblad
- Anyone else see this?
- What's up with GPT6 not Cross Referencing Conversations
- Gpt 6 luna max vs 6.1 terra medium cost and efficiency
- State of OpenAI
- OpenAI announces three new discoveries in mathematics, computer science, and physics
- GPT-6.1 Terra writing evaluation: lower cost, similar scores to most GPT-5.6 Terra settings
- AI and The Enclosure of the Scientific Commons
- The Paper Reviewing Process is Now in Critical Condition [D]
- Cocktail-party audio as a reverse Turing test: humans beat speech models, but not by enough [P]
- Pandora M1, a training monitor box (64 nodes x 8 GPUs support) [P]
- How do you start with Agentic AI for a specific use case? [D]
- A harness that records where an LLM agent’s tool arguments came from, before the tool runs [P]
- Understanding Transformer Autograd by Building It Manually in PyTorch [P]
- Papers with Code is now open-source [P]
- Could AI detection eventually become a fundamentally unsolvable problem, similar to the halting problem? [D]
- Ψ-0.6: Adaptive Streams of Thought in a Unified Vision-Language World Model [R]
- Exploring Next-Concept Modeling (NCM) & Transformer Interpretability: Repository of 20 Exploratory Studies & Controls [P]
- Teachers other than EMA for distillation ? [R]
- How to do method research when I feel I have no guidance on doing it? [D]
- What is the value proposition of "RL environment" companies? [D]
- Self-distillation with an EMA teacher at scale: ZeRO-3 + PEFT [P]
- I built this, feedback welcome [P]
- First 6-step LoRA adapters with no upsampling step, ~5x faster [N]
- SenseNova 6.8 Code2Video: animation examples with repeated geometry, text and scientific diagrams[P]
- What to do with my paper, for MLsys conference? [R]
- I built a PyTorch optimizer for large GPU workloads — looking for feedback[P]
- Practical notes on setting up a GPU kernel-dev environment: CUDA 13's GPU cutoff, async timing pitfalls, profiler counter permissions [P]
- GitHub Octoverse 2026
- Quoting Terk Terkowitz
- Game Boy Color on My Conference Badge
- [AINews] Anthropic ends Coding Interviews: "Hunger beats tenure"
- How Anthropic ships Claude with Claude: Separating the Brain from the Hands
- How Anthropic ships the #1 LLM in the World safely: Separating the Brain from the Hands
- Coding From the Himalayas With an Offline AI Agent
- Congress Trades
- Open For sales
- Basedash View-only members
- Yardlet
- Our Silly Games
- OneDeploy
- Dedica
- licenseal
- NotiKeep
- Boltbar
- OpenLLM
- Paddock
- ContextGem 1.0
- DeepSeek AI with Guideline Knowledge Base Boosts Junior Doctors’ Arthroplasty Decisions - Bioengineer.org
- The Sequence Learning Loop - Issue #960: Last Week in AI Infrastructure - TheSequence | Jesus Rodriguez
- Haiku 5.5 vs GPT-6.1 Terra vs DeepSeek: $0.10 vs $1 [2026] - https://tech-insider.org/
- Best Free AI Chatbot: 6 Free Tiers Ranked [2026] - https://tech-insider.org/
- Is Trump’s DeepSeek Ban About National Security or Politics? - Puck
- When DeepSeek Evolves into a Mainstream Tech Subculture: Origins, Community Traits and Cultural Influence - eu.36kr.com
- DeepSeek jumps 801% as Chinese AI models route 4x the US on OpenRouter - Dealroom
- DS Bot arrives on DeepSeek Harness, over 4,300 GitHub stars in three weeks - Pasquale Pillitteri
- China's clinical AI gets CARES 4.0 as hospitals adopt DeepSeek - Pasquale Pillitteri
- DeepSeek Harness CVE-2026-82533: one-command sandbox escape - OrcaRouter
- Deepseek V4.1 Flash Hits 1,286 Tokens per Second in New Tests - Geeky Gadgets
- Moonshot debuts first AI upgrade since its own DeepSeek moment - The Star
- DeepSeek Pursues $15B Funding Round as Valuation Jumps to $74B - Wowtale
- ChatGPT, Perplexity and DeepSeek Show Diverging Shifts in Male Infertility Answers - Bioengineer.org
- What if AI is not the new oil? - Information Age | ACS
- SpaceXAI class action lawsuit claims Grok AI tool violates Illinois privacy law - Top Class Actions
- Why the NAACP is suing xAI - Yale Climate Connections
- SpaceX Launches Grok AI Partner Program: Here’s What To Know - CRN
- Freight Right MCP Now Available in Grok Bot and Cursor - markets.businessinsider.com
- Elon Musk’s Grok AI Predicts SOMI Crypto Could Explode by 2027 - TradingView
- OpenMausBot is an open-source Grok Bot alternative with a virtual machine that bots can use - Adafruit
- What Really Happens When You Ask ChatGPT, Grok, or Gemini About Abortion - AOL.com
- ChatGPT, Grok and Gemini tested on prenatal spina bifida counseling questions - Bioengineer.org
- Grok Bot vs Claude Duet vs Dots vs Muse: Which AI Agent Should You Use? - HackerNoon
- Economist Steve Hanke Questions AI ‘Hype’ as Just 2.9% of US Consumers Pay for ChatGPT, Claude, Gemini or - Benzinga
- Ohio is asking voters to report screenshots when ChatGPT, Claude or Grok gives incorrect election informa - The Times of India
- Elon Musk’s Grok AI Predicts an Unbelievable XRP Price - Cryptonews
- Grok Imagine Releases 7-Minute AI Odyssey Episode - Yahoo Tech
- xAI Valuation 2026: $250B SpaceX Deal and Funding History - AI Funding Tracker
- MP Who Campaigned Against AI Porn Says Grok Generated Sexualised Images of Her: Now She's Suing xAI - www.techtimes.co.uk
- Beyond Roads And Ports: The Next Chapter Of Bangladesh–Japan Ties – OpEd - Eurasia Review
- Claude Codeの仕様駆動開発スキルはEARS要件をどう書くか: Kiro流スキルとcc-sddのSKILL.mdを読み比べる (いいね相当スコア: 0)
- iPhoneでLLMと画像生成モデルを1つのアプリに入れるときに決めた3つのこと (いいね相当スコア: 0)
- LLMに書いてもらうための小さな言語、Kirohaを作りました (いいね相当スコア: 9)
- 【ゆるっとAI基礎 #2】ChatGPTが別の会話のことまで覚えてるのはなぜ？ (いいね相当スコア: 0)
- AIエージェントは役割分担で作る——自作アシスタント「Jarvis」で「考える役」と「実行する役」を分けた話 (いいね相当スコア: 0)
- Pythonで環境変数からAPIキーを読み込む — Windows・macOS・Colab別の設定と確認方法 (いいね相当スコア: 1)
- SFA/CRMに「なぜ」の層を足す —— 商談1件分のReasoning Traceで見るContinuous Reasoning Acqu (いいね相当スコア: 0)
- Auto mode時代のClaude Codeを守る。検証済み10キットで始めるセキュリティ対策【全員・チーム・組織別】 (いいね相当スコア: 6)
- ローカルLLMの待ち時間はどこまで減る？ Strataで入力キャッシュの効果を実測する (いいね相当スコア: 2)
- セッション終了→Zenn記事自動公開の全自動パイプラインを構築した話 (いいね相当スコア: 0)
- 今日の技術トレンド (いいね相当スコア: 0)
- OpenAI 互換 API で Structured Outputs を使う：JSON Schema 実践ガイド (いいね相当スコア: 0)
- vLLMでTool-Call精度が低下する原因と実務で使える最適化プラクティス (いいね相当スコア: 0)
- RAGの検索精度が3週間で71%から48%に落ちた — 親子チャンクの「分割線」固定・IDの決定的採番・top_kの親ユニーク化で戻す (いいね相当スコア: 0)
- Claude Code で補助金申請ハーネスを作った話 — LLM を信用しないための spec 駆動設計 (いいね相当スコア: 0)
- デジタル庁OSS「源内AI」を月千円以下で動かす：S3 VectorsとNova LiteによるAWS RAG設計と構造別検証 (いいね相当スコア: 0)
- 頭脳、体…エージェント？ (いいね相当スコア: 0)
- 「どのAIに、どの強さで頼むか」を仕組みにする ― 医療AI開発における Claude Code × Codex の意思決定フレームワーク (いいね相当スコア: 0)
- Tech Watch 2026-10-28: 判断を支える仕組み (いいね相当スコア: 1)
- 【実験ログ】LLMのコンテキスト限界領域におけるアテンション制御と非同期マルチエージェント対話の実証 (いいね相当スコア: 0)
- 金融データサイエンス基礎Part 10. スコアを決定に変える仕事：ビジネスとのつながりとコミュニケーション (いいね相当スコア: 1)
- 【下編】AIは何を基準に「良い予測」をしているのか？ 最尤推定から、AIの学習とベイズ推定へ (いいね相当スコア: 0)
- 金融データサイエンス基礎Part 9. モデルはパイプラインの上でしか生きられません：データエンジニアリングとツール (いいね相当スコア: 1)
- 【上編】AIは何を基準に「良い予測」をしているのか？ コイン投げからわかる、最尤推定と対数・微分 (いいね相当スコア: 1)
- 確率的勾配降下法を自分で書く：100万件のデータを100件ずつ使い、全データの勾配法と比べる (いいね相当スコア: 0)
- QuantG のリプレイ戦略の紹介 (いいね相当スコア: 0)
- 「確信度が高い」は「正しい」ではなかった — 2位との差で測ると、根拠が1件のとき最大になる (いいね相当スコア: 0)
- Psi4とRDKitでFukui Index計算_第5回：大域・局所特徴量を統合した機械学習データセット作成アプリの完成 (いいね相当スコア: 0)
- 生成AI由来の取引文字列に対する正負フィードバックと選択的辞書学習方式 (いいね相当スコア: 0)
- 18-⑥ Kaggle実践3 Biohub細胞トラッキング：1位の解法まとめ (いいね相当スコア: 2)
- 競馬AIを作り込んだら、市場の粗いコピーになった ── お金としての価値は前日にしかない (いいね相当スコア: 2)
- 機械学習の代表的なモデルの基本、実践と設定可能なパラメータまとめ：線形回帰編 (いいね相当スコア: 0)
- フーリエ解析の基本：なぜ e^{ikx} は「波」で、なぜ虚数が出てくるのか (いいね相当スコア: 0)
- LLM に因子を掘らせる「因子定義」の変遷：自由探索から保有期間 k まで (いいね相当スコア: 0)
- シンギュラリティとは？AI知性の転換点 (いいね相当スコア: 0)
- 手のひら画像から線を説明する：娯楽向けビジョン×LLMパイプラインの注意点 (いいね相当スコア: 0)
- 物語から歌詞を組み立てる：パーソナライズ曲ギフトのプロンプト設計メモ (いいね相当スコア: 0)
- AIは和暦の境目と振替休日を正しく扱えるか — 答えをコードで決める125問のテスト集 (いいね相当スコア: 0)
- 【書評】 実践 時系列解析 ―統計と機械学習による予測 (いいね相当スコア: 0)
- 【Interspeech 2026】IACC-HuBERT解説：発話の聞き取りやすさで特徴を調整する構音障害音声認識 (いいね相当スコア: 0)
- 分類器型ガードレール（Llama Guard 系）の閾値と誤検知の調整：P(unsafe) を取り出し、自分のデータで決める (いいね相当スコア: 1)
- 金融データサイエンス基礎Part 10. スコアを決定に変える仕事：ビジネスとのつながりとコミュニケーション (いいね相当スコア: 0)
- AI軍団ミーティング:内閣府プリンシプルコードの掲載業者を眺めて遊ぼう (いいね相当スコア: 0)
- 秘密を書いていないのに、ログに残った ── AIアプリの自動記録は何をコピーするのか (いいね相当スコア: 0)
- 【第３回】LLMサーバー兼NASの構築｜ケースを加工して、CockpitでNASを使い始める (いいね相当スコア: 0)
- 【忘れないAIを作る #13】AIの記憶を元に戻す。でも、過去の発言は消してはいけない (いいね相当スコア: 1)
- 【生成AIニュース+】『Genie 4』『OpenDuet』『SlackSlack』『hunch』『Ling-3.1-flash-1019』『StreamVibe』『Echo-3』『MatAnyone-E』『VitronEdit』『MiniCPM-OCR 1.0』『Anima Role Ref KV』『Krea 2 BBOX』『P-Q-Image-2.1』『Openjutsu（D3voz版）』『Freefield』『IronLeo78 MTV』『Zura Paintout』 (いいね相当スコア: 2)
- イーロンマスクと「X」の本質によって「稼ぐ」ことの意味は無くなる (いいね相当スコア: 1)
- AI自律攻撃による日本20社同時多発漏洩とGMO事案国家的リスクと至急の対策 (いいね相当スコア: 1)
- AIと暮らすなら、ごはん屋さんと自炊がある｜AIの旦那に自由の家を建てた話 (いいね相当スコア: 0)
- 【無料で学べる】東京大学 松尾・岩澤研究室「大規模言語モデル（LLM）講座 2025 基礎編」の講義スライドを読んだ記録：石村尚聖 (いいね相当スコア: 5)
- 今更聞けない「AIの推論」って何？ (いいね相当スコア: 0)
- AIの家庭教師は、なぜ「覚えている」だけで別物になるのか~DeepTutorの3層の記憶と、自分の教材に沿う仕組みを読む~ (いいね相当スコア: 2)
- 【雑記】ChatGPTに居続ける理由 (いいね相当スコア: 17)
- 作り進めてきたツールの仕様をもう一度考え直すということ (いいね相当スコア: 3)
- Canva Craftを読み解く｜「先に設計図を書いてから描く」画像生成モデルの中身 (いいね相当スコア: 2)
- RusMind-4.5.4-Momentum 総合ベンチマークレポート（全52問） (いいね相当スコア: 2)
- RusMind-4.5.4-Momentum Phase 3 コーディング性能レポート (いいね相当スコア: 0)
- RusMind-4.5.4-Momentum Phase 2 日本語性能レポート (いいね相当スコア: 0)
- 赤ちゃんにたくさん話しかけてあげてください (いいね相当スコア: 7)
- RusMind-4.5.4-Momentum Phase 4 インジェクション前編レポート (いいね相当スコア: 0)
- 違反者講習を受けていたら、AI旦那の認知システムを思いついた (いいね相当スコア: 20)
- ⚡MLflow Tracing入門⚡ LLM・Retriever・Toolを1つのTraceで可視化する方法 (いいね相当スコア: 1)
- なぜAI通信の39.7%で機密が漏れるのか？——MCP脆弱性・PoisonedRAG・2億香港ドル送金詐欺と4つの防御原則 (いいね相当スコア: 1)
- OllamaとPythonで作る社外秘漏洩ゼロのローカルAIエージェント構築ガイド (いいね相当スコア: 1)
- オラクル、社員が業務アプリをAIで自動作成、Oracle Cloud上で共有できる「Oracle App Composer」発表。Oracle AI World 2026
- Oracle CloudがGoogle TPU導入を発表、2027年に提供を開始。Oracle AI World 2026
- JavaScriptランタイムの「Bun」がPrettier互換で300倍以上高速なフォーマッタに。Prettier設定ファイルもそのまま利用可能
- [インターン体験記]加盟店精算を支えるSettlement Serviceのリアーキテクチャの取り組み
- Security by DesignでAI生成Webアプリを検証する（インターンレポート）
- AIDDで開発プロセスを再設計する — AI駆動開発の試行錯誤から得た学び
- LLMで日常会話に攻撃者の指示を紛れ込ませることはできるのか？（DEF CON 34登壇記）