AI News Digest 2026-09-17
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
参考記事一覧を表示
- ローカルAIを選ぶ前に、確かめたい5つのこと [importance:0 dev:38] (いいね相当スコア: 0)
- Gemini Live audio [importance:75 dev:72]
- Claude Cowork and chat are now one Claude [importance:62 dev:45]
- Anthropic, OpenAI, and xAI Want to Slow Down Artificial Intelligence (AI) Development. These 2 Stocks Could Be the Biggest Losers. - The Motley Fool [importance:28 dev:15]
- Elon Musk, the world’s richest man, says he’s living in an Airstream trailer to oversee xAI’s biggest expansion yet - Fortune [importance:12 dev:5]
- As the world debates the risks of AI, China closes the technology gap with the US - thebusinessjournal.com [importance:25 dev:10]
- 高速な判断に特化したAI - TypeSafe「Jev」 と System One Model [importance:65 dev:72] (いいね相当スコア: 取得失敗)
- Small Programming Tricks [importance:48 dev:75]
- Dream-RSI: Recursive Self-Improvement through Evolving Worlds [importance:52 dev:78]
- Mistral X Mozilla: Private, Multilingual AI Browsing [importance:68 dev:72]
- Tell the speakers that you liked their talks [importance:5 dev:10]
- Show HN: An e-ink frame that hears birds and draws them as 1800s illustrations [importance:35 dev:58]
- The DeepMind Institute [importance:72 dev:52]
- How big are factorials? [importance:2 dev:5]
- Apple Reference Image: A New Approach for Verified Photography [importance:48 dev:58]
- The Siberian Ice Maiden and the Scythian World [importance:0 dev:0]
- Learning Programming in an Age of LLMs [importance:55 dev:72]
- GitHub Is Having Trouble Counting Things [importance:22 dev:38]
- Can we stop with the uptime percentages? [importance:20 dev:42]
- Kyber (YC W23) Is Hiring a Forward Deployed Engineer [importance:8 dev:28]
- Show HN: How Stale Is Your AI? Release age and training cutoff for 20 models [importance:68 dev:75]
- ER visits for gambling disorders doubled after expanded online gambling market [importance:2 dev:2]
- A warning about 'model welfare' [importance:45 dev:48]
- Anatomy of a Texture [importance:15 dev:52]
- This Code Is CRAP (2011) [importance:5 dev:35]
- Original Sony PlayStation 2 security chip 'broken wide open' after 26 years [importance:28 dev:42]
- Scaling Golang CI by Replacing actions/setup-go [importance:55 dev:82]
- Why I'm still bearish on LLMs after Navier-Stokes [importance:62 dev:75]
- v2.1.273 [importance:58 dev:85]
- v2.1.272 [importance:35 dev:78]
- Helping older adults use AI in everyday life [importance:15 dev:8]
- Reimagining advertising with AI [importance:35 dev:45]
- How workers are unlocking new ways of working [importance:28 dev:25]
- AI for Societal Impact [importance:12 dev:20]
- Building AI to accelerate science and improve lives [importance:8 dev:15]
- AI for everyone in every language [importance:8 dev:15]
- New insights from Google’s AI & Economy ATLAS [importance:12 dev:25]
- Bypassing inference bottlenecks: Accelerating complex AI search with Retrieve-for-Train [importance:72 dev:82]
- Your Agent Aced the Task. Will It Do It Again? [importance:12 dev:32]
- Have it both ways: stay discoverable in search while disallowing AI training [importance:55 dev:72]
- Give every teammate and agent the right level of access to your Workers [importance:62 dev:80]
- Logpoints Walkthrough [importance:32 dev:78]
- Rider and ReSharper 2026.2.2 Are Out! [importance:48 dev:78]
- Behind the Scenes: How the OpenTelemetry Plugin Maps Your Microservices in Real-Time [importance:58 dev:82]
- IntelliJ IDEA 2026.2.3 Is Out! [importance:28 dev:72]
- Java 27 in IntelliJ IDEA [importance:35 dev:72]
- Translating CUDA Tile Operations from Python to Rust Using Agentic AI [importance:78 dev:88]
- Dense vs. MoE Models: Active Parameters, Throughput, and When to Choose Each [importance:68 dev:78]
- How NVIDIA Groq 3 LPX Deterministic Execution Drives Power-Efficient High-Interactivity Inference on NVIDIA Vera Rubin [importance:62 dev:78]
- How NVIDIA NVLink 6 Delivers Multi-Layer Resiliency for AI Factories [importance:55 dev:75]
- Scaling Federated Learning Across Docker, Kubernetes, and Slurm with NVIDIA FLARE [importance:65 dev:82]
- Building the materials foundation for AI [importance:48 dev:35]
- Roundtables: Could AI really kill us all? [importance:38 dev:25]
- What’s at stake in AI’s trillion-dollar gamble [importance:32 dev:12]
- Agility’s new humanoid robot will stop, squat to avoid harming human coworkers [importance:28 dev:32]
- Exclusive: Paying for frontier AI models buys 4-month head start at 5x the cost [importance:58 dev:52]
- AI labs want in-house auditors — but maybe they should shut the front door first [importance:32 dev:28]
- Your AI agents can now control your Google Home devices [importance:68 dev:82]
- Robots are waiting for a ChatGPT moment: Nvidia’s Les Karpas explains why at TechCrunch Disrupt 2026 [importance:15 dev:22]
- Threads’ new features let podcasters promote shows and reach listeners [importance:5 dev:8]
- Next wave of VCs judging Startup Battlefield 200 contenders at TechCrunch Disrupt 2026 revealed [importance:5 dev:5]
- 3 days left to exhibit: Get your brand in front of VCs and high-value leads at TechCrunch Disrupt 2026 [importance:2 dev:2]
- SK Hynix reportedly in talks with Intel to build memory chips in US [importance:32 dev:25]
- Former Infosys chief’s AI startup nabs another $53M [importance:25 dev:18]
- Amazon launches Alexa+ in India with Hindi support [importance:22 dev:22]
- We don’t need AI regulation — leave safety to us, Nvidia’s Jensen Huang says [importance:18 dev:8]
- The AI data center boom is colliding with cities scarred by big industry [importance:28 dev:8]
- Meta now lets AI agents handle the boring parts of WhatsApp Business setup [importance:72 dev:85]
- The AI graveyard: a running list of projects and startups that didn’t make it [importance:28 dev:12]
- US data centers could consume more natural gas than Germany and Japan combined by 2035 [importance:32 dev:12]
- AI agents now have a place to snitch [importance:42 dev:48]
- Meta expands subscription push with new AI-focused plans [importance:32 dev:12]
- OpenAI, Anthropic, Google have been in talks on AI safety for weeks [importance:48 dev:35]
- AEO startup Profound hits unicorn valuation, raises $180M Series D 7 months after last round [importance:22 dev:8]
- Former TikTok execs built an app that uses AI to teach you how to pose for a photo [importance:18 dev:22]
- Apple might make servers again to cash in on the AI rush [importance:52 dev:48]
- Google will now let any AI agent run your smart home [importance:68 dev:82]
- Claude comes for Gemini with its own take on Docs and Slides [importance:65 dev:62]
- The sexy AI-powered dating app scams are here [importance:25 dev:15]
- A brief history of AI executives calling for regulation [importance:28 dev:12]
- AI and data centers are incredibly unpopular in every poll [importance:25 dev:8]
- Meta’s new One subscriptions put a price on social media and AI [importance:32 dev:15]
- This doorbell camera lets a human security guard watch your front door [importance:15 dev:8]
- Is Big Tech’s AI slowdown a safety pact or a cartel? [importance:35 dev:15]
- What execs and politicians are saying about slowing down AI development [importance:28 dev:12]
- Dropbox Evolves Riviera Content Processing Platform to Support AI Workloads [importance:68 dev:82]
- Article: Your Next DSL Author Is a Language Model [importance:72 dev:85]
- Presentation: Teaching Engineers, Trusting AI: How Education Enabled Autonomous Code Review [importance:68 dev:82]
- Dropbox Outlines How Focusing on Existing Infrastructure Efficiency Can Create Headroom for AI [importance:65 dev:80]
- Grab's Agent Framework LLM-Kit Accelerates AI Agent Production Deployment [importance:72 dev:85]
- Rethinking Robot Safety in the Age of AI [importance:48 dev:35]
- EU president warns AI agents "escaping their environment" are just a preview of what's coming [importance:42 dev:32]
- Apple is reportedly building an enterprise AI server with its own M8 Ultra chips [importance:55 dev:52]
- Google Deepmind launches interdisciplinary institute to tackle the big questions around AGI [importance:68 dev:48]
- Former OpenAI researcher builds an AI model that judges options instead of writing text [importance:72 dev:82]
- Mozilla's new Smart Window assistant runs on Mistral's models [importance:68 dev:78]
- Political opposites unite in Washington to rein in AI [importance:28 dev:12]
- Nearly one in five AI researchers already expected an extinction scenario from AI back in 2024 [importance:52 dev:25]
- Google launches Gemini 3.8 Live to take on OpenAI's GPT-Live-1 at a fraction of the cost [importance:75 dev:72]
- AI labs have a data trust problem that their policies haven't solved [importance:48 dev:28]
- ZGCM-1: A Fully Open and Extremely Efficient Foundation Model for Math and Agentic Search [importance:50 dev:65]
- Converge Then Diversify: Decoupling Convergence and Diversity in Multi-Objective Bayesian Optimisation [importance:25 dev:50]
- Generalized Agent Iteration: One Formal Framework for Iterative Policy Improvement and Recursive Self-Improvement [importance:45 dev:70]
- Vibe Patenting: Evaluating LLM Judges for Professional Patent-Drafting Agents [importance:35 dev:60]
- Toward Self-Adaptive Physical AI: Can LLM Agents Manage Long-Horizon Physical Tasks? [importance:50 dev:60]
- LabAgent: Customize Any Research Hubs for Scientific Discoveries Using AI Agents [importance:45 dev:65]
- TimeThink: Eliciting Compositional Reasoning in Timeseries Large Language Models [importance:35 dev:50]
- Root-Cause Attribution Is a Search Problem: Continual Search for Long-Horizon Agent Failures [importance:55 dev:75]
- Governing at Machine Speed: An Adaptive Intelligence Architecture for Real-Time AI Policy Enforcement [importance:60 dev:65]
- OrchSLM: Probing the Dynamics of Small Language Model Orchestration [importance:60 dev:75]
- Grounded Adjudication of Variations across Extracted TimeLines (GAVEL): Comparing Clinical Timelines Against Their Case Reports [importance:35 dev:45]
- Token Efficient Task Execution via Application Behavior Modeling for Web Agents [importance:55 dev:70]
- Fraglingo: Molecular Design via Attachment-Aware Autoregressive Fragment Generation [importance:30 dev:40]
- From Legal Text to AI-specific Risk Sources: A Systematic Analysis of the EU AI Act's High-Risk Requirements [importance:65 dev:60]
- Asclepius: An Adaptive Harness for Long-Horizon Clinical Agents [importance:50 dev:70]
- AutoTailor: Automatic, User-Aligned Capability Selection and Adaptation for Web Agents [importance:60 dev:75]
- Toward a Decision-Assurance Layer for AI-Assisted Flight Planning in Air Traffic Management [importance:50 dev:65]
- Carbon-Aware Routing for Function Calling in Edge-Cloud LLM Systems [importance:50 dev:70]
- A Hybrid Agentic AI Framework for Intelligent Supply Chain Analytics [importance:45 dev:65]
- Planning or Learning: Reliability and Cost in Multi-Asset Maintenance [importance:40 dev:55]
- Causal multi-modal AI for personalized chemosensitivity prediction [importance:35 dev:45]
- How User-AI Mistreatment Occurs and Matters in Conversational Systems? [importance:45 dev:60]
- FLoKD: Adaptive Knowledge Distillation for Federated Low-Rank LLM over Wireless Networks [importance:50 dev:70]
- $\tau$-Elicitation: Benchmarking multi-turn entity extraction in voice agents [importance:45 dev:70]
- Identity Is More Than Recall: A Benchmark for Persistent Identity in Deployed AI Agents [importance:55 dev:75]
- Solar Intelligence [importance:40 dev:60]
- Safety as a Constraint: Fine-Tuning a LLM Recommender to Explain Itself [importance:45 dev:60]
- FedV-KGQA in Practice: Design Lessons and an Interactive Prototype [importance:35 dev:60]
- Cost Characterization of Vertically Partitioned Federated Knowledge Graphs [importance:30 dev:55]
- GeoSkill:Experience-Driven Hierarchical Skill Learning with Collaborative Revision forGeospatialAgents [importance:50 dev:70]
- Enhancing Event Candidate Acquisition for Event Linking [importance:40 dev:60]
- Recoverability as a System Primitive for Long-Horizon AI Agents [importance:60 dev:75]
- Windowed A-K-MDP [importance:25 dev:50]
- Drift-Constrained Optimization: Only Direction Matters in Fine-Tuning Instruct Models [importance:50 dev:70]
- Degraded but Not Entirely Ineffective: PE-Based Deformable Graph Neural Networks [importance:30 dev:55]
- JaxAHT: A JAX-Based Library for Ad Hoc Teamwork [importance:50 dev:75]
- MANAS-2: Constrained Reconstruction for EEG Foundation Models [importance:35 dev:50]
- IBBench-Light: A Paired Evaluation of Task-Conditioned Responses to External Directives [importance:40 dev:65]
- Trustworthy Agentic AI: A Comprehensive Cybersecurity and Systems Survey on Threat Landscapes, Defense Architectures, and Open Challenges [importance:70 dev:80]
- How Many Thoughts Can a Vector Hold? The Capacity of Reasoning by Superposition [importance:45 dev:65]
- Positioning manuscripts in the scientific landscape with agentic AI [importance:40 dev:60]
- Homeostatic Continual Learning [importance:45 dev:65]
- Surprising Effectiveness of Self-Demonstrations in Enhancing Schema-Ontology Mapping with LLMs [importance:40 dev:60]
- Partition Scores Are Not System Scores: Deployment-Fidelity Gaps in Decomposed Algorithm Selection [importance:35 dev:60]
- Do Not Restart: Residual Completion for Stateful Agent Handoffs [importance:55 dev:75]
- LLM-Enhanced Multi-Agent Reinforcement Learning for Unified Electric Vehicles-Charging Station-Grid Optimization in Public Charging Systems [importance:40 dev:60]
- Bypass Observation: A Conceptual Design of a Non-Intrusive Layer-Wise Semantic Extraction Architecture [importance:45 dev:65]
- ViperQ: Order Flow Pattern Recognition via Auction Market Theory for Reinforcement Learning Trading [importance:35 dev:55]
- UniCAR-RL: Seeing Better before Thinking Deeper in Visual Mathematics [importance:45 dev:65]
- ClinAgent: A ReAct-Based Agent for Conversational Access to Clinical Trial Information [importance:40 dev:65]
- Map Users and Mapmakers: The Scope of Cognitive Attribution from Acquired Representations [importance:30 dev:50]
- LoRA Fine-Tuned Models for Control Systems Course Q\&A: A Multidimensional Evaluation of Model Scale and Rank Effects [importance:35 dev:60]
- SAILOR: Solver-Assisted Interactive LLM-based Optimization Recovery [importance:50 dev:70]
- Synthetic Data in Marketing Research: How to Evaluate and When to Trust [importance:40 dev:55]
- Convergent Emergence of In-Context Learning Across Modalities [importance:55 dev:70]
- Schizophrenia Detection from EEG Signals: A Transformer Framework with Spectrogram Representation [importance:30 dev:50]
- VeriDx: Earning the Right to Diagnose with Disease-Centric Verification [importance:50 dev:65]
- Semantic Knowledge Technologies: what the Semantic Web lost sight of, and what it never had [importance:40 dev:55]
- MOSCOPT: Mixture-of-Skills Collective Optimization for LLM Agents [importance:55 dev:75]
- Dynamic Learning Solutions: A System for Personalized Educational Video Generation [importance:40 dev:60]
- Question's Gambit: The First Move Matters in Agentic Deep Search [importance:50 dev:70]
- Safety Signals to Verify NetOps Agents with Action-Level Granularity [importance:55 dev:75]
- When does a scaling result justify a different allocation? A critical review of resource-allocation evidence for AI systems [importance:50 dev:70]
- Beyond Scene Description: Multi-Agent Orchestration for Non-visual Access to Virtual Worlds [importance:50 dev:70]
- OptoAgent: A Trustworthy Multi-Agent Framework for Opportunistic Vision Micro-Screening in Classroom Environments [importance:40 dev:65]
- AI Deployment Accountability Engineering: A Vision for Accountable AI in Safety-Critical Socio-Technical Systems [importance:60 dev:75]
- A note on goal-based hierarchical RL [importance:35 dev:60]
- DynSTEER: Dynamic Stage-wise Trajectory Evaluation and Execution-time Review for Agents [importance:55 dev:75]
- Diffusion-Based Generation of Gait Trajectories [importance:35 dev:55]
- Lightning Weave: Improving the Accuracy-Efficiency Frontier of Reasoning Models through Capability Composition [importance:55 dev:75]
- Depth and Scale in the Sub-150M Regime: JugnuLM-53M vs JugnuLM-110M [importance:45 dev:70]
- Moral Rebel Agents: Decision-Making Under Conflicting Obligations [importance:40 dev:65]
- Bayesian Intelligence from the Outside [importance:35 dev:55]
- AppliedScientist: Automated Scientific Revision Through Iterative AI Reviewing [importance:50 dev:70]
- AcquireBound: Runtime Authorization for Resources Acquired by AI Agents [importance:60 dev:75]
- AI Persuasion as a Threat to Human Control [importance:60 dev:70]
- Another Blueprint In The Wall: How to Ask Frontier AI Like a Kid? [importance:45 dev:70]
- Crypto Accounting Bench: Evaluating Frontier and Open-Weight Models on Crypto-Asset Accounting Tasks [importance:35 dev:55]
- ANASSA: An Agentic AI Orchestration Framework for Spatial Intelligence [importance:50 dev:75]
- One Model, Two Physical Stories: Auditing Misalignment in Multi-Modal World Modeling [importance:45 dev:65]
- El Agente Potente: High-Throughput Agentic Atomistic Simulations [importance:50 dev:70]
- Self-Orchestrating Language Models: Leveraging Semantic Dependence for Efficient Inference [importance:55 dev:75]
- Domain Generalization for Smartphone-Based Human Activity Recognition: A Systematic Analysis of Components and Interactions [importance:35 dev:55]
- GGUF-Metadata Prediction of Single-Sequence llama.cpp Throughput Across Three Systems [importance:45 dev:75]
- Externalizing Requirement-to-Repair Artifacts as Observable Traces for LLM-Based Program Repair [importance:55 dev:80]
- Geometric Flow enhanced Graph Coarsening [importance:30 dev:55]
- Towards a knowledge-enhanced single-cell foundation model [importance:40 dev:60]
- MemRiskBench: Trace-Aware Risk-Preserving Evaluation for Long-Horizon LLM Agents [importance:60 dev:75]
- Converting Sequenced Fuzzy Cognitive Maps to Causal Virtual Worlds with Large Video Generators [importance:45 dev:65]
- Shallow Beliefs: Synthetic document finetuning does not inoculate against emergent misalignment from reward hacking [importance:55 dev:70]
- CoMem: Collective-Individual Memory Synergy for Evolutionary Multi-Agent Systems [importance:55 dev:75]
- Semantic-TVM: Structure-Preserving Trustworthy Virtual Memory for Memory-Augmented and Tool-Using Agents [importance:60 dev:75]
- Overflip: Repetition-Induced Label Flips in Guardrail Models [importance:50 dev:70]
- Four Ledgers, Not One Score: Responsible Communication of LLM-Judge Calibration in Biomedical ML [importance:50 dev:65]
- Horizon-specific Expert Fusion for Photovoltaic Power Forecasting [importance:35 dev:55]
- The average-farmer illusion in language-model simulations of agricultural decisions [importance:40 dev:60]
- BusMA: A Bus Communication Substrate for Multi-Agent Systems [importance:55 dev:75]
- Enabling Creative Exploration for Vibe Design Agents [importance:45 dev:70]
- ER-EDF: A Psychology-Grounded Emotion Regulation Framework for Speech Empathetic Dialogue Generation in Large Audio-Language Models [importance:40 dev:65]
- OpenAI4S: Code as Action, Science as Sessions [importance:55 dev:75]
- Medical Knowledge Simplification for Patients in the Era of LLMs: A Case Study on Diabetes [importance:30 dev:25]
- HazardAuditor: From Executable Threats to Safer Computer-Use Agents [importance:55 dev:70]
- T-LoopFormer: Token-Level Elastic-Depth Looped Transformers for Latent Reasoning With Dynamic Routing [importance:45 dev:80]
- STHMoE: Hypergraph-Enhanced Heterogeneous Dependency Coordination for LLM-Based Urban Traffic Data Forecasting [importance:35 dev:40]
- VisInteract: Towards Dynamic Interactive Text-to-Visualization under Imperfect Queries [importance:40 dev:70]
- Issue Bias in Generative AI Writing Assistance: Political Issues and LLMs in the Swedish 2026 Election [importance:25 dev:30]
- From Ideas to Actions: A Public-Data Decision-Support Toolchain Across the Venture Lifecycle [importance:35 dev:55]
- CWM: Controllable White-Box Meta-Prompting for Adaptive Retrieval-Augmented Generation and Reasoning Ability [importance:50 dev:75]
- Empirical Evaluation of Open-Source Large Language Models for Retrieval-Augmented Generation in ESG Domain [importance:40 dev:65]
- ProIQA: A Process-Based Framework for Fine-Grained Math Item Quality Assessment [importance:35 dev:55]
- Why LLM Agents Collapse Without Oversight: The Enforcement Gap as the Mechanism Behind Emergence World Failures [importance:60 dev:75]
- Reason What Matters: Retrieval-Grounded Reasoning for Universal Multimodal Embeddings [importance:45 dev:75]
- Evaluation Metrics for Safe Reinforcement Learning [importance:50 dev:75]
- Parameter-Efficient Adaptation of Pretrained Language Models for Time-Series Forecasting [importance:45 dev:75]
- MAPS: Memory-Aware Predictive Scheduling Framework for Large Language Model Serving [importance:60 dev:85]
- RSIAgent: Autonomous Exploration for Recursive Self-improvement in New Environments [importance:55 dev:80]
- SkillLift: Learning Dense Rubrics from Sparse Oracles for Efficient Skill Evolution [importance:55 dev:80]
- When Tool Calls Succeed but Workflows Fail: Anomalies at the Agent-Tool Boundary [importance:60 dev:85]
- Who Teaches Which Token? Verifier-Gated Multi-Expert On-Policy Distillation for Scientific Reasoning [importance:50 dev:75]
- Can AI systems have free will? [importance:20 dev:10]
- Empirical Evaluation of Task-Based Permission Scoping Architecture for AI Agents [importance:60 dev:85]
- HISPO: Hierarchical Importance-Sampling Policy Optimization with Entropy-Derived Segments [importance:50 dev:75]
- The Troy Moment of AI: Why Some Will Cheat and Some Will Follow? [importance:65 dev:75]
- Beyond Safe Answers: Segment-Aware Listwise Alignment for Reasoning Safety in Large Reasoning Models [importance:55 dev:80]
- Option-Aware Retrieval and Task-Specific VLM Adaptation for Medical VQA [importance:40 dev:70]
- GRIN+: Towards Fast Yet Effective Machine Unlearning for Imbalanced Medical Data [importance:50 dev:75]
- Diversified and Perceptible Counterfactual Examples Leveraging Expert Knowledge [importance:40 dev:70]
- Potential of Artificial Intelligence Algorithms for Identification of Relevant Diagnostic and Prognostic Biomarkers of Early-Stage Liver Cancer [importance:35 dev:40]
- EEG-Xplain: Decoding Neural Black-Boxes of EEG Foundation Models [importance:45 dev:70]
- NoteVQA: Benchmarking VLMs on Real-Life Questions from Human Communities [importance:45 dev:75]
- Beyond Accuracy: Robustness, Cost, and Governance Trade-offs for Vision-Language Models in Templated Document Extraction [importance:50 dev:75]
- New Conditions for Philosophers to Catch the Wave of Citizen Deliberation in the Age of Artificial Intelligence in advance [importance:20 dev:15]
- Predicting build orientation for SLM dental parts: a comparison of rotation representations and direct vector regression [importance:25 dev:50]
- Data storytelling meets interpretable machine learning: Decoding AI decisions for non-experts without revealing sensitive data and model details [importance:40 dev:65]
- Are LLMs Good Financial User Simulators? A Preliminary Study [importance:35 dev:55]
- Design of a Deep Learning Credit Risk Early Warning System Integrating Multi-source Heterogeneous Data [importance:40 dev:65]
- EvoOntology: A Self-Evolving Ontology Layer for Data Agents [importance:55 dev:85]
- KnowBench: Effort Reduction as a Unified, Deployment-Grounded Benchmark for Clinical AI [importance:50 dev:70]
- Navigating Sparse Evidence: Agentic Visual RAG via Explicit Context Selection and Consolidation [importance:55 dev:80]
- When Should a World Model Move? Loss-Conditioned State Execution [importance:45 dev:75]
- Atria Dawn: The Dawn of Agentic Superintelligence [importance:80 dev:90]
- AlgoEvo: Self-Evolving Agentic Search for Automated Algorithm Discovery [importance:55 dev:85]
- LongAgent: History-Guided Agentic Search for Longitudinal Outcome Prediction [importance:50 dev:75]
- Pilot Early, Commit Late: A Real-Options Model of Enterprise AI Adoption under Rapid Technological Progress [importance:35 dev:30]
- Recurrent GraphNeural NetworkswithSet-BasedAggregation [importance:40 dev:75]
- Stellar Colosseum: A Many-Agent Harness for Long-Horizon Research in Mathematics and Theoretical Computer Science [importance:60 dev:85]
- Corrupt Plans, Clean Traces: Evading Chain-of-Thought Monitoring with Plan Injection [importance:65 dev:85]
- Towards Optimizing SQL Generation via LLM Routing [importance:50 dev:80]
- Beyond Quacking: Deep Integration of Language Models and RAG into DuckDB [importance:60 dev:90]
- Generalization Can Emerge in Tabular Foundation Models From a Single Table [importance:55 dev:80]
- Physically Aware Radiomics Without Interpolation: Disentangling Voxel Geometry and Signal Modification in CT and MRI [importance:40 dev:50]
- Natural-Language to SysMLv2 Translation via Conformance-Driven Iterative Refinement [importance:50 dev:80]
- Multilingual Agent System for Inclusive Wildfire Evacuation Guidance [importance:50 dev:70]
- From Semantic to Token Communication: The Next Paradigm for Large-Model-Driven 6G Intelligent Connectivity [importance:50 dev:75]
- Is Bash All You Need? An Empirical Study of Tool Interfaces for Enterprise Digital Worker Agents [importance:60 dev:85]
- LLMs or Naive Bayes? Old Gems or New Ways [importance:50 dev:80]
- Diagnosing Faults in Reinforcement Learning Simulators and World Models with Canonical Polynomial Invariants [importance:50 dev:80]
- Planning as Dynamics Relaxation: Hippocampal Recurrent Network Realizes Optimal Goal-Directed Navigation [importance:40 dev:50]
- Chemical and geometric representation fidelity improves drug--target affinity prediction [importance:45 dev:65]
- ShieldVLA: Feasibility-Aware Safety Alignment for Vision-Language-Action Models [importance:55 dev:80]
- Evaluation of MLLM-Agnostic Plug-and-Play Keyframe Selection Methods for Long Video Understanding [importance:45 dev:75]
- An Evolutionary Computation Framework for Multi-Agent Q-Learning with Mean-Field Environmental Feedback [importance:50 dev:80]
- (How) Do MLLMs Report Bistable Images Like Humans? [importance:40 dev:70]
- Sampling headroom is not selection gain: a compute-value audit of test-time scaling for video world models [importance:50 dev:80]
- TryOnReward: Learning Foveated Consistency for Reinforcement Fine-Tuning of Virtual Try-On [importance:40 dev:70]
- From Process Loss to Assembly Bonus: Human-Grounded Diagnosis of Multi-Agent LLM Collaboration [importance:55 dev:80]
- BEACON: Behavior and Appearance Control for Subject-Specific Video Generation [importance:50 dev:75]
- Forward-Facing Near-Infrared Adds Little to Colour for Farm-Machinery Traversability: A Site-Disjoint Evaluation of Sensor-Dependent Spatial Leakage [importance:35 dev:60]
- Conflict-Predictive Variable Horizons in Multi-Drone Distributed Model Predictive Control [importance:45 dev:75]
- Multimodal-Multiresolution Foundation Model for Lunar Remote Sensing [importance:55 dev:80]
- LLaDA-UI: Bringing Block-wise Diffusion to Vision-Language GUI Agents [importance:60 dev:85]
- Variational Template Matching with Statistical Fusion for Anomaly Detection in Patterned Structures [importance:40 dev:70]
- Adaptive Conformal Redistribution for Inter-class Transitional Uncertainty in Medical Image Classification [importance:40 dev:65]
- IMM-based Multiple Object Tracking using a State Prediction Neural Network [importance:50 dev:75]
- Task-Based CT Protocol Optimization Using Reinforcement Learning and Virtual Imaging Trials [importance:45 dev:70]
- Feasibility and Memory Mechanisms of Chern-Simons Context Reservoir Computation [importance:35 dev:70]
- Decoupling Error Attribution in Cloud-Native Graph-RAG: A Data Integrity Diagnostic Framework [importance:55 dev:85]
- Pedestrian Crossing Intent Classification From Event-Based Vision Using Convolutional Spiking Neural Networks With Temporal Augmentation [importance:50 dev:75]
- The Agentic Company OS: Substrate Inversion for Sustained Enterprise Agent Deployment [importance:65 dev:90]
- Bridging Thought and Action: Taming Long-Horizon Instability in Open-Source LLM Agents with a MetaTool-Enhanced ROS Framework [importance:60 dev:85]
- ProtoCAM: Interpretable Few-Shot Mask-Guided Prototypical Learning for Breast Lesion Classification in Ultrasound Imaging [importance:40 dev:65]
- SkillAtlas: An Attack Trace Library for Agent Skills [importance:60 dev:85]
- Task-Aware Federated Fine-Tuning for MoE-based Large Language Models [importance:55 dev:85]
- ReCAST: Reward Credit Assignment across Timesteps for Online Diffusion Reinforcement [importance:55 dev:85]
- Chance-Constrained Belief-Space Maneuver Planning for Autonomous Collision Avoidance Under Uncertainty [importance:50 dev:75]
- Certifiably Interpretable Training of ReLU-MLPs for Boolean Tasks with Guaranteed Truth-Table Generalization [importance:50 dev:80]
- Learning to Solve Hard Problems in RL for LLMs by Never Giving Up [importance:55 dev:80]
- Hindsight Bias in Clinical Temporal Reasoning: How Future Data Exposure Affects Large Language Model Judgment [importance:50 dev:70]
- Symmetry- and Property-Aware Crystal Generation with Reinforcement Learning for Inverse Materials Design [importance:50 dev:75]
- Edge-addition monotonicity of positive p-energy fails for every p >= 1 [importance:30 dev:40]
- Generative AI and Extended Reality in Collaborative Architectural Design Education: An Exploratory Studio Study [importance:35 dev:55]
- Building a Production Greek-English Speech Recognizer [importance:55 dev:85]
- Canaries in the Bank: Auditing User-Level Privacy in Private Evolution [importance:55 dev:85]
- One Spectrum, Two Resources: Data-Memory Scaling in Autoregressive Prediction [importance:45 dev:80]
- A Three-Axis Stress Test of LLM vs Classical ML for Network Intrusion Detection under Distribution Shift and Adversarial Evasion [importance:55 dev:85]
- Adaptive Phase-Switching for Communication-Efficient Federated LoRA Fine-Tuning [importance:55 dev:85]
- Positive Topology and Feasible Refinement: Forcing Matrices, Positivity, and Information [importance:35 dev:60]
- Rolling Day-Wise Mortality Prediction in Critically Ill Patients With AKI on CRRT Utilizing Machine Pressure Waveforms [importance:45 dev:65]
- Generative Interpretability via Scalable Neuro-Symbolic Models [importance:55 dev:85]
- Attention Is All You Need (to Avoid Spurious Oscillations) [importance:45 dev:75]
- Domain-Specific Jargon in Large Language Models: A Comparative Analysis between General-Purpose and Specialist Models [importance:25 dev:35]
- Same Patient, Different Order: Action-Level Reliability of Clinical LLM Agents Under Repeated Runs [importance:35 dev:50]
- mKernel: Fast Multi-GPU, Multi-Node Fused Kernels [importance:70 dev:75]
- Predictive audio representations for early detection and tracking of hidden dynamic objects [importance:50 dev:55]
- An Efficient and Modular Framework for Targeted Harm Mitigation in LLMS [importance:55 dev:65]
- FaithfulBench: Does AI Counsel Uphold or Undermine the User's Professed Faith? [importance:20 dev:25]
- LayerRoute: Adaptive Layer-Skipping with LoRA-Preserved Quality for Efficient LLM Inference [importance:70 dev:75]
- Not all Negation Cues are Equal: Affixal Negations Yield Better Negation Understanding [importance:35 dev:55]
- Leakage-Safe and Scheduler-Aware Machine Learning for Grid Job Runtime Prediction [importance:45 dev:60]
- Gap Entropy and Almost Instance-Wise Optimal Best-Arm Identification [importance:20 dev:30]
- Oops, Not Now: PEARL, a RAG-Based Support Agent for Gameplay and What Players Want from AI Help [importance:45 dev:60]
- TyPatch: Transforming Patches into Typestate Rules for Kernel Bug Detection [importance:55 dev:70]
- PolicyMem: Geometric Policy Memory for LLM Governance [importance:65 dev:75]
- HarnessBandit: Joint Learnability-Transferability Scheduling for Multi-Harness Agentic Reinforcement Learning [importance:65 dev:75]
- On the Equivalence of Stochastic Control and Path Space Formulations for Schr\"odinger Bridges over Compact Connected Lie Groups [importance:15 dev:20]
- GEAR: From Dynamic Encoding to Dynamic Activation in Social Trajectory Prediction [importance:45 dev:55]
- Understanding the Limits of Agentic ICD Coding [importance:35 dev:50]
- Exploring Automated Vulnerability Identification in JavaScript Code Using Large Language Models [importance:65 dev:75]
- CRAF: Cross-View Residual-Aware Fusion for Deepfake Speech Detection [importance:50 dev:65]
- LePlanner: An Iterative Amortized Controller For World Models [importance:60 dev:70]
- ReWeight: Leveraging Human Data for VLA Post-Training via Demonstration Retrieval and Sample Weighting [importance:65 dev:75]
- Bangla Sentence Function Classification: Corpus Development, Model Benchmarking, and Interpretability [importance:30 dev:50]
- Trustworthy, Explainable, and Sustainable Decentralized Intelligence for 6G Networks [importance:50 dev:65]
- When Malicious Instructions Persist: Persistent Memory Poisoning Attack on Harness-Based Agents [importance:70 dev:80]
- Phorecaster365: A Human-Supervised Reference Architecture for Hybrid Pharmaceutical Sales Forecasting and Planning Decision Support [importance:35 dev:40]
- DiTAR+: Dual Optimization for Robust Autoregressive Diffusion Speech Synthesis [importance:60 dev:75]
- Finite-Time Node Separation in Recurrent Graph Neural Networks with Persistent Gaussian Perturbations [importance:40 dev:65]
- CRITICS - Critical Science Without Borders: Language Models to Promote Critical Thinking in Science Education [importance:30 dev:50]
- Hardware-Aware Learned Representation Compression for Distributed In-Sensor Vision [importance:55 dev:70]
- Thought without systematicity? Evaluating reasoning models on rule induction tasks [importance:50 dev:60]
- SGWIB:Sliced Gromov-Wasserstein Information Bottleneck for Video Highlight Detection [importance:45 dev:60]
- Mizan: A National Benchmark for Evaluating Large Language Models on Iraqi Arabic and the Iraqi Civic Context [importance:25 dev:45]
- Rethinking the Implications of Human Feedback for Preference Learning in Human-Robot Collaboration [importance:50 dev:65]
- Confuse the Model, Control the Flow: Understanding and Mitigating Privacy Leakage from LLM Agents with Information Flow Control [importance:75 dev:80]
- AGENTQ: Quantization-Conditioned Backdoor Attacks on LLM Agents [importance:75 dev:80]
- GraMRAG: Orchestrating Multi-Agent Multi-Step Reasoning via Graph Memory with Reinforcement Learning [importance:65 dev:75]
- LPA-CWM: A Learned Physical Adjudicator for Motion Reasoning with Counterfactual World Models [importance:45 dev:65]
- RA-CoA: Training-free Fashion Image Captioning via Retrieval-Augmented Chain-of-Attributes [importance:40 dev:55]
- A Voxel-Spacing-Aware Extension of PyRadiomics for Anisotropic Texture Analysis [importance:30 dev:50]
- Real-Time Synthesis of Robust Controlled Invariant Sets for Monotone Systems [importance:50 dev:65]
- Talking to Me or Someone Else? Rethinking Talk-to-Me Detection in Egocentric Videos [importance:35 dev:50]
- To do($x$) or not to do($x$): Medical Image Counterfactuals for Dataset Augmentation [importance:35 dev:50]
- LIMBO: Lifelong Inference-Time Memory and Budget Optimization for LLM Agents [importance:75 dev:80]
- A New Transformer-Based Approach for Audio-Based Kinship Verification and a New Uncontrolled Mandarin Kinship Speech Dataset [importance:25 dev:45]
- Signatures of Steerability in Activation Space of Language Models [importance:65 dev:75]
- Towards Evolving Context Parameterization for Large Language Models [importance:60 dev:75]
- Bi-Level Routing and Sparse Spatial Attention based Multi-View BEV 3D Object Detection for Autonomous Driving [importance:55 dev:70]
- Entropy-Punctured Bloom Filters for Memory-Efficient Machine Learning [importance:50 dev:70]
- Data-free On-policy Distillation [importance:70 dev:80]
- ECAS: An Edge-Controlled Agentic System for Validation-Gated Scientific Application Execution [importance:70 dev:80]
- Task-Specified Active Metrological Inspection with Measurement-Steered VLA Manipulation and Deterministic Evidence Gating [importance:55 dev:70]
- Modeling, Scaling, and Decoding: Optimizing Controllable Speech Generation with Nonverbal Vocalizations [importance:50 dev:70]
- Graph-Transformer Fraud Detection with Self-Supervised Pretraining and Conformal Risk Control [importance:60 dev:75]
- Assessing the Applicability of Existing Design Recommendations to AI Companion Design: A Multi-Method Study [importance:35 dev:45]
- OpWeave: Flexible Operator Disaggregation for Heterogeneous LLM Serving [importance:70 dev:80]
- The Attribution-Compression Frontier in Retrieval-Augmented Generation [importance:65 dev:75]
- ATTRICITE: Training an Open 4B Model for Citation Recovery toward Faithful Attribution [importance:60 dev:75]
- SpermYOLO: A Coordinated YOLO-Based Detector for Accurate and Efficient Sperm and Impurity Detection in Microscopic Images [importance:20 dev:45]
- Biquaternionic Space with Complex-valued Attention for Temporal Knowledge Graph Completion [importance:45 dev:65]
- AURA: Unified Multimodal Framework for Conversational Music Editing [importance:40 dev:60]
- LLaTSA: Large Language Model-Aligned General-Purpose Transient Stability Analysis [importance:55 dev:70]
- A Hybrid Dependency-Aware Framework for Task Decomposition and Dynamic Agent Generation in Oracle-to-PostgreSQL Migration [importance:65 dev:80]
- Surrogate-Assisted Genetic Programming with Phenotypic Characterisation in Dynamic Multi-Mode Project Scheduling [importance:40 dev:60]
- A Generative AI Integrated Multimodal Framework for Low-Latency Multi-Camera Person Re-Identification [importance:50 dev:65]
- Has Scientific Talent Shifted from Depth to Breadth?Evidence across Papers, Knowledge Inputs, Careers, and Teams [importance:40 dev:50]
- Lightweight Generalized DeepFake Face Detection with WAVIE: Wavelet Augmented Vision Intermediate Embeddings [importance:55 dev:70]
- A latent dimension of Condorcet's jury theorem for multiple AI advisers [importance:35 dev:60]
- From Visual Attribution to Clinical Reasoning: Explainable Parkinson's Disease Screening from Hand-Drawn Patterns [importance:35 dev:50]
- NeuroActiSep: Detecting Factual Hallucinations from Feed-Forward Neurons in a Single Pass [importance:70 dev:80]
- Follow the Geometry, Not the Model: Cold Start Semi-Supervised Learning [importance:50 dev:70]
- Bridging the Modality Gap in Long-Form Clinical Audio: A Comparative Study of Lightweight and Heavyweight End-to-End SOAP Generation [importance:40 dev:60]
- EdgeHAR: An Edge-Native Compact Sensor Foundation Model for Human Activity Recognition [importance:60 dev:75]
- Sharing standardized image-derived data in computational pathology using DICOM [importance:30 dev:55]
- Proving olympiad geometry theorems on a superconducting quantum processor [importance:50 dev:70]
- Disentangling Topology and Diversity in Multi-Agent LLMs for Multilingual Low-Resource Emotion Detection [importance:60 dev:75]
- AlgoRAG: Retrieval-Augmented Generation for Theoretical Computer Science Education -- A Comprehensive Evaluation Framework for Algorithm Analysis and Complexity Theory [importance:45 dev:65]
- SENTINEL: A Multi-Pathway Architecture for Detecting Living-Off-the-Land APT Attacks on Windows Command Lines [importance:70 dev:80]
- Diagnosing Temporal Misalignment in Multichannel Time-Series Classification with Minimum Description Length [importance:45 dev:65]
- Open-UniMo: Towards Unified Motion-Language Understanding and Generation in the Open World [importance:55 dev:75]
- Investigating the Impacts of Generative AI on Information Seeking [importance:35 dev:35]
- CompCQR: Compositional Query Generation for Training-Free Conversational Search [importance:60 dev:75]
- Skill Composition for Legged Robot Reinforcement Learning [importance:55 dev:75]
- Natural Language Knowledge Graph Query Execution: Leveraging Controlled Semantics in the LLM Context Window [importance:60 dev:75]
- Compositional SVG Generation via VLM-Driven Hierarchical Semantic Parsing [importance:55 dev:75]
- MANE: A Multi-Path Adaptive Network for Edge Onloading of Deep Neural Networks [importance:60 dev:75]
- PU classification under Non-SCAR: clustering-assisted logistic model with oversampling enhancement [importance:35 dev:60]
- WaterKron and FlipFlop Hessian: Information-Theoretically Grounded Quantization with Kronecker-factored Hessians [importance:65 dev:80]
- Carryover Drafting: Recycling Rejected States for Speculative Decoding [importance:70 dev:85]
- OCT-FedSIR: Toward Trustworthy Federated Ophthalmic Learning under Annotation Noise [importance:40 dev:65]
- Building Legal Reward Models for Grounding and Abstention [importance:60 dev:75]
- A property-registry contract for retrieve-or-refuse thermal-mechanical lattice search [importance:35 dev:55]
- Calibrating Interpretability Instruments Before Trusting Their Verdicts [importance:65 dev:80]
- Refusal Reads Only a Slice of What the Model Knows: Harm-Keyed Routing and Its Exceptions Across Model Families [importance:70 dev:80]
- From Visual Feedback to Textual Reviews: A Multi-Agent Vision-Language Framework for Image-Grounded Review Assistance [importance:45 dev:65]
- TriCalRAG: A Three-Strategy, Retrieval-Augmented Benchmark for On-Premise LLM-Based Root Cause Analysis in AIOps [importance:70 dev:80]
- Loop-Back Authority in LLM Agent Teams: A Paired Experiment on Flat and Hierarchical Coordination [importance:65 dev:80]
- How broad is that claim? Mapping Generalisation in NLP Research [importance:40 dev:60]
- The Stochastic Deputy: Structural Tenant Isolation for Tool-Using LLM Agents [importance:75 dev:85]
- Mind Which Bird You Favour: Parameterizing Adequacy-Fluency Balance in Meta-Evaluation of Machine Translation [importance:40 dev:60]
- A primer on evaluation methods for large language models in healthcare [importance:50 dev:65]
- Route, Don't Fix: Regime-Dependent Decoding Correction and a Trajectory-Gated Router for Reliable Clinical LLM Answer Selection [importance:45 dev:65]
- Enemray: Toward Capable Language Models for Hassaniya [importance:35 dev:70]
- Efficiency Hallucination: Formalizing and Measuring Behavioral Calibration in LLM-Based Code Optimization [importance:55 dev:75]
- A Responsive Present, a Shared Past, a Social Other: Teens' Overreliance on Companion AI Chatbots [importance:30 dev:15]
- LLMs as Oracles: Reliance on LLMs for Subjective Personal Questions [importance:25 dev:10]
- RAIN: Region-Aware Inversion Network for Semantic Watermark Extraction [importance:50 dev:70]
- One Example Is Enough to Pass Fairness Benchmarks: Rethinking Fairness Evaluation for Aligned LLMs [importance:60 dev:75]
- Semantic Fibers and Cross-Gram Interference: A Calculus of Safety Drift in Overcomplete Representations [importance:55 dev:75]
- Interpolation Is Not Invariance: Pair Count Is Not Coverage in Transformation Audits [importance:40 dev:65]
- SeqMaestro: From nucleotide sequences to biological hypotheses through interpretable machine learning [importance:35 dev:50]
- PeerPen: AI-Assisted Writing for Online Mental Health Peer Support [importance:35 dev:35]
- Forty Shades of Blue: Quality-Diversity Alignment via Mode-Conditioned Reinforcement Learning [importance:65 dev:75]
- Neural-Network Solutions to Real-Space Charge Density and Generalization [importance:30 dev:50]
- Cross-Block Conditioning in Deep Boltzmann Machines for Statistical Data Fusion [importance:30 dev:60]
- CAL-MOS: Bridging Layers with Adapters for Robust MOS Prediction Across Speech Foundation Models [importance:45 dev:70]
- Online Language Adaptive Sampling for Better Distributed Cross-lingual Gains [importance:50 dev:75]
- LiftGCN: Efficient Energy-Preserving Graph Learning via Joukowski Spectral Lifting for Finite Element Stress Prediction [importance:35 dev:65]
- ActGuard: Pre-execution Action Auditing against Indirect Prompt Injection in LLM Agents [importance:70 dev:85]
- IMPACT-VLA: Interaction-aware Multimodal Propagation Attribution via Counterfactual Trajectories for Vision-Language-Action Policies [importance:45 dev:70]
- Steering Generative Robot Policies with Lexicographic Preferences [importance:50 dev:70]
- PIDS-Bench: Evaluating Prompt-Injection Detectors Under Over-Defense, Obfuscation, and Distribution Shift [importance:65 dev:80]
- Pick Your Poison: Learning to Select Poison Sets for Stronger LLM Backdoor Attacks [importance:70 dev:80]
- Validating Hybrid-State Cache Recovery for GLM-5.3-Flash with vLLM and LMCache [importance:50 dev:80]
- SpliTEE: Improving LLM Inference on Trusted Hardware with Differentially Private GPU Outsourcing [importance:65 dev:80]
- Mirror, Mirror on the Wall: Prompt Echoing in Small Instruct Language Models [importance:40 dev:65]
- Personalizing Personal Health Interfaces: Co-Design with Generative AI [importance:35 dev:40]
- Ensemble Complexity in Photovoltaic Forecasting [importance:35 dev:55]
- Not All Prompts Are Equal: Exploration-Guided Prompt Scaffolding for Multimodal Reinforcement Post-Training [importance:55 dev:75]
- Salesforce Koa: An Enterprise Language Model for Agentic Tool Use [importance:80 dev:85]
- Rethinking Procedural Audio Pre-training: Source Scaling and Objective Adaptation [importance:45 dev:70]
- Branched Optimal Transport Amortization [importance:40 dev:70]
- Translating the Translator: Decomposing the Cost of English-Forced Inter-Agent Communication [importance:60 dev:80]
- Beyond Numerical Time Series: A Unified Benchmark for Multimodal Forecasting with Heterogeneous Context [importance:50 dev:75]
- Generate to Explore, Select to Exploit: Aligning LLM-based Headline Generation with Personalized Recommendation [importance:50 dev:70]
- ChatGPT Images 2.5 in the Wild: A Launch-Period Dataset and Detector Evaluation [importance:45 dev:70]
- Physics Informed Neural Network model for the dynamical study of Abdominal Aortic Aneurysm [importance:35 dev:60]
- Legislating World-Model-Based Planning with Legal Reasoning [importance:40 dev:65]
- DepthBenchCAD: When Does Deeper Auditing Yield More Reliable Conclusions? [importance:50 dev:75]
- Refinement-based Flow Policy Optimization [importance:50 dev:75]
- MoME: Mixture-of-Memory Embeddings for Context-Aware Sparse Lookup [importance:60 dev:80]
- AdaVSkip: Adaptive Visual Token Skipping Across Layers For Efficient MLLMs Inference [importance:65 dev:80]
- PACE: Progressive Angular-to-Norm Contrastive Embedding [importance:50 dev:75]
- EMR: Self-Evolving Medical Multi-Agent System via Experience Mining and Reuse [importance:60 dev:80]
- Interpreting hierarchical organisation of speaker embeddings [importance:40 dev:70]
- TEAR: Table Extraction with Attribute Recommendation from Texts via Large Language Models [importance:50 dev:75]
- Failure-Guided Co-Evolution of Prompts and Training Data [importance:65 dev:80]
- Augmenting Large Audio-Language Models with Frame-Level Grounding for Fine-Grained Temporal Perception [importance:55 dev:75]
- Pre-PEFT Probing: Weight Statistics and Perturbation Robustness for Layer Selection in VLM Vision Encoders [importance:55 dev:80]
- ProtoGuide: Prototype-Driven Guidance for Class-Conditional Graph Generation [importance:50 dev:75]
- Math for AI safety: an invitation for mathematicians [importance:65 dev:70]
- When Correlations Mislead: Confounder-Aware Multi-View Urban Region Representation Learning [importance:40 dev:70]
- The Universe of Universes: Benefit Yield Functions, Implosion Thresholds, and Infrastructure-Aware Optimization in Multi-LLM Systems [importance:75 dev:85]
- Clean Scores, Buried Evidence, and Confident Wrong: A Receipt-Based Audit of Frontier Agentic QA [importance:70 dev:80]
- Planning in the Backbone: DiffAdapterVLA for Native Continuous Trajectory Generation with Driving VLMs [importance:60 dev:75]
- Concept-Grounded Reasoning with Prompt-Driven Localization for Interpretable Structured Report Generation [importance:55 dev:75]
- Dynamic Semantic Compression for Efficient Latent-Space Inference in Large Language Models [importance:70 dev:85]
- End-to-End Cell Detection via Instance-aware Graph Modeling [importance:45 dev:70]
- Robust and Efficient Communication for Multi-Agent Learning [importance:65 dev:80]
- Divide, Consult, Conquer: Capability Laundering Through Aligned LLMs [importance:75 dev:80]
- IWC-Bench: Evaluating Web Application Generation from a Software Testing Perspective [importance:65 dev:85]
- CodeTS: Verifiable Text-to-Time Series Generation via Executable Code [importance:55 dev:80]
- A Conservative OCR-Enabled Workflow for R214 Sodium Screening of South African Packaged Foods [importance:30 dev:55]
- On the role of the tokenizer in ECG transformer models [importance:40 dev:70]
- Turkish MMLU Pro: Traceable Option Augmentation and Its Validity Limits in Turkish Multiple-Choice Evaluation [importance:50 dev:75]
- Spook the Machine: Gamified Exploration of Human Imagination of Machine Fear [importance:30 dev:30]
- How Lossless Is Lossless Speculative Decoding? The Role of Numerical Precision in Orthrus [importance:65 dev:85]
- Authorship attribution and aesthetic evaluation of AI poetry: a case study with Haiku [importance:25 dev:35]
- Automating Attack Graph Construction for Agentic Pentesting. Towards Neuro-Symbolic Vulnerability Hunting [importance:70 dev:85]
- The Misery of Mechanistic Interpretability: A Formal Perspective [importance:55 dev:75]
- Specifying Reward Functions for RL Without Environment Sampling [importance:55 dev:75]
- Don't Count the Edits, Judge by the Outcome Alone: Reward-Based Evaluation for Grammatical Error Correction [importance:50 dev:75]
- PIVOT: Physics-Grounded Verification for AI-Generated Audio-Video Detection [importance:65 dev:75]
- Big Brains and Changing Environments: Cause or Consequence? [importance:20 dev:20]
- Self-Evolving Memory for Generative Recommendation [importance:60 dev:75]
- A Unified Vision-Language Model for PSMA PET/CT Report Generation, Visual Question Answering, and Lesion Segmentation [importance:55 dev:75]
- VideoScout: Learning Agentic Active Exploration with Adaptive Reasoning Pacing for Long Video Understanding [importance:65 dev:80]
- Through the Eyes of the Beholder: Biometric and Demographic Conditioning for Multimodal Sexism Detection [importance:45 dev:65]
- Multi-View Molecular Representation Learning with Hierarchical Graphs and Contextualized Fingerprints [importance:40 dev:70]
- Beyond AI Literacy: A Structured Review and Exploratory Meta-Analysis of Measures for Competent Generative-AI Use [importance:55 dev:60]
- FedLTLib: A Comprehensive Benchmark for Federated Long-Tail Learning [importance:60 dev:80]
- ModaLens: Measuring Image Sensitivity in Report-Conditioned Medical VLMs [importance:50 dev:75]
- Predictive Likelihood Ratios for Language Model Watermark Detection [importance:55 dev:75]
- KaiNinja: Extending Native 3D Generators to the Part Level [importance:50 dev:70]
- CIDERS: Cloud-Edge LLM Collaborative Learning via Accelerating Personalized Bilevel Optimization [importance:65 dev:85]
- Circuit-MLLM: Topological Logic-Guided Latent-Space Visual Reasoning for Circuit Schematic Understanding [importance:55 dev:75]
- Benchmarking Intra-Patient 3D Deformable Multimodal Image Registration [importance:45 dev:70]
- Don't Send What You Don't Need: Question-Guided Token Pruning as a Privacy Defense for Vision-Language Models [importance:70 dev:85]
- Scalability and Performance Evaluation of Federated Learning Frameworks: A Comparative Analysis [importance:60 dev:85]
- More Than Just Access: Generative AI as Communication Intermediary for Blind and Low-Vision Users [importance:50 dev:60]
- Bench2Dex: Benchmarking Visuo-Tactile Bimanual Dexterous Manipulation Across Dexterous Hands [importance:55 dev:75]
- A Language-Guided Multimodal Foundation Model for Zero-Shot and Multi-Task Brain Signal Analysis [importance:55 dev:75]
- Look Before You Leap: Factual Decoding with Internal Attribution Signals [importance:70 dev:85]
- Sylvas: Synergistic Learning Value based Device Scheduling in Federated Continual Learning [importance:60 dev:80]
- Event-Native Symbolic-Temporal Spike Encoding Framework for Heterogeneous Cyber Streams [importance:50 dev:75]
- Transfer Learning for Socioeconomic Estimation in Forced-Displacement Settings [importance:40 dev:65]
- When the World Lies: Backdoor Attacks on Latent World Models for Downstream Control [importance:75 dev:85]
- Delegating Authorization to Misaligned Agents: Coalitional Alignment and Safe Control [importance:75 dev:85]
- CiteGuard-RAG: A Validation-Centered AI System for Evidence-Grounded Question Answering [importance:70 dev:85]
- Per-Matrix Optimality Is Not Enough: Three-Level Optimization for Low-Rank LLM Compression [importance:70 dev:85]
- Before You Poll with LLMs: A Deliberative Diagnostic Framework [importance:55 dev:75]
- K-Bench: a clinically calibrated benchmark for evaluating large language models in high-risk mental health conversations [importance:30 dev:50]
- LLM-Based Schema-Aware Split Learning for Privacy-Preserving Mental Distress Prediction Across Heterogeneous Surveys [importance:25 dev:40]
- Learning Multimodal One-step Flow Policy via Value-weighted Optimal Transport [importance:25 dev:30]
- Privacy-enhanced federated learning via asynchronous aggregation and local differential perturbation [importance:30 dev:50]
- Anatomical Grounding and Leakage-Aware Multimodal Contrastive Learning for Alzheimer's Disease Classification from Structural MRI [importance:20 dev:35]
- SlipSense: Multimodal Tactile Learning for Low-Latency and Generalized Slip Detection [importance:25 dev:35]
- Vulnerability Localization Benchmark: Measuring Agentic Security Analysis at Repository Scale [importance:55 dev:70]
- The Router Within: Eliciting Native Skill Routing from a Frozen LLM [importance:55 dev:75]
- Estimating Uncertain Spatial Relationships in Robotics [importance:20 dev:40]
- FICAug: Feature-Informed Clustering and Augmentation for Facial-Expression-Based Parkinson's Disease Screening [importance:15 dev:30]
- Hallucination in Multimodal Foundation Models: A Survey on Causes, Corrections, and Evaluations [importance:50 dev:60]
- Efficient On-Device Agents via Adaptive Context Management [importance:60 dev:75]
- DeepFeature: LLM-Empowered Context-aware Feature Generation for Wearable Biosignals [importance:30 dev:45]
- Privileged observations enable rapid and reliable policy discovery directly in the physical world [importance:25 dev:35]
- Echo-CoPilot: A Multiple-Perspective Agentic Framework for Reliable Echocardiography Interpretation [importance:30 dev:55]
- MCPAgentBench: A Real-world Task Benchmark for Evaluating LLM Agent MCP Tool Use [importance:70 dev:85]
- ClinicalReTrial: Clinical Trial Redesign with Self-Evolving Agents [importance:25 dev:50]
- Towards a Mechanistic Understanding of Propositional Logical Reasoning in Large Language Models [importance:45 dev:60]
- Modality-Guided Mixture of Structured Experts with Entropy-Triggered Routing for Multimodal Recommendation [importance:30 dev:55]
- Tool Use Reduces Depth-Induced Collapse in OOD Reasoning [importance:50 dev:65]
- NeuroProlog: Multi-Task Fine-Tuning for Neurosymbolic Mathematical Reasoning via the Cocktail Effect [importance:40 dev:65]
- TimeWarp: Evaluating Web Agents by Revisiting the Past [importance:60 dev:75]
- From Refusal Tokens to Refusal Control: Discovering and Steering Category-Specific Refusal Directions [importance:50 dev:70]
- vla-eval: A Unified Evaluation Harness for Vision-Language-Action Models [importance:55 dev:80]
- ZEBRAARENA: A Diagnostic Simulation Environment for Studying Reasoning-Action Coupling in Tool-Augmented LLMs [importance:55 dev:75]
- Ventriloquist LLMs: Linear Alignment of Late-Stage Representations [importance:35 dev:60]
- Utility-Guided Agent Orchestration for Efficient LLM Tool Use [importance:60 dev:75]
- MARCUS: An agentic, multimodal vision-language model for cardiac diagnosis and management [importance:30 dev:55]
- Mecha-nudges for Machines [importance:45 dev:60]
- Auditable Agents [importance:55 dev:70]
- CLEAR: Context Augmentation from Contrastive Learning of Experience via Agentic Reflection [importance:55 dev:75]
- Multi-Agent Empowerment and Emergence of Complex Behavior in Groups [importance:35 dev:60]
- Iterative Multimodal Retrieval-Augmented Generation for Medical Question Answering [importance:35 dev:60]
- Valley3: Scaling Omni Foundation Models for E-commerce [importance:55 dev:70]
- Post-Reasoning: Improving the Performance of Non-Thinking Models at No Cost [importance:50 dev:70]
- Personality engineering with AI agents: A new methodology for negotiation research [importance:40 dev:55]
- Playing Devil's Advocate: Off-the-Shelf Persona Vectors Rival Targeted Steering for Sycophancy [importance:45 dev:65]
- Counteraction-Aware Multi-Teacher On-Policy Distillation for General Capability Recovery with Domain Preservation [importance:40 dev:70]
- FundaPod: A Multi-Persona Agent Pod Architecture with Knowledge Graph Memory for AI-Assisted Fundamental Investment Research [importance:50 dev:70]
- PortBench: A Correlation-Aware, Full-Pipeline Benchmark for LLM-Driven Portfolio Management [importance:45 dev:60]
- The Theory of Mind Utility: A Formal Account of Mentalizing [importance:35 dev:50]
- WISE: A Long-Horizon Agent in Minecraft with Why-Which Reasoning [importance:40 dev:65]
- FinAcumen: Financial Multimodal Reasoning via Self-Evolving Experience Memory Harness [importance:50 dev:75]
- Data Scale, Not Latency, Shapes Cross-Lingual Encoder Transfer in Streaming ASR [importance:30 dev:55]
- Humans Disengage, Reasoning Models Persist: Separating Difficulty Registration from Deliberation Allocation [importance:45 dev:65]
- CLQT: A Closed-Loop, Cost-Aware, Strategy-Consistent Benchmark for Diagnostic Evaluation of LLM Portfolio-Management Agents [importance:50 dev:75]
- Spatial Reasoning via Modality Switching Between Language and Symbolic Representations [importance:45 dev:70]
- SportD: How do VLMs physically strategize? [importance:35 dev:60]
- Design and Embedded Validation of Compact ML Models for Affective Touch Classification in a Soft Interactive Companion [importance:25 dev:50]
- TRACTA: Benchmarking Temporal Reasoning over Semantic Trajectories [importance:40 dev:65]
- Evolving from Lessons: Skill-Augmented Table Graph Reasoning for Operation-wise Table Question Answering [importance:40 dev:65]
- Hidden APIs in Language Models: Discovering Reusable Causal Interfaces from Forked Futures [importance:45 dev:70]
- Whetstones: Measuring Coevolution Between Adaptive Malware and Behavioral Defense [importance:35 dev:55]
- Modeling Social Dynamics with an LLM-Enabled Agent Based Network-Dynamic (LAND) Model [importance:40 dev:65]
- Self-Certification of Representation Adequacy: Sequential Certification at Minimum Task Loss [importance:35 dev:60]
- ChartAnno: Benchmarking Multimodal Large Language Models for Chart Annotation Generation [importance:40 dev:65]
- Negotiating Risk Boundaries in AI for Policing Through Mixed-Stakeholder Deliberation [importance:40 dev:45]
- An Explainable GNN Framework for Component-Level Anomaly Diagnosis [importance:35 dev:65]
- Hierarchical Compositionality for An Assistive AI Agent [importance:45 dev:75]
- Measuring Cross-Task Behavioral Consistency in Language Model Agents [importance:50 dev:75]
- Development and Feasibility Evaluation of an Edge AI as Medical Device System for Breast Cancer Multidisciplinary Team Meetings [importance:30 dev:60]
- Autonomous Mathematical Discovery in an Open-World Multi-Agent Environment [importance:50 dev:75]
- A Contract-Centered Architecture for Scalable and Manageable Agentic Runtimes [importance:70 dev:85]
- Responsible Integration of AI in Cancer Genomics: Barriers, Risks, and Pathways to Trustworthy Clinical Translation [importance:25 dev:50]
- Reconciling Process Supervision with Outcome-Based Credit in Agentic Policy Optimization [importance:55 dev:75]
- Verifiable Disaster Storylines and Causal Knowledge Graphs: A Citation-Grounded Pipeline from Heterogeneous Humanitarian Sources [importance:40 dev:60]
- Bioinfoysis Technical Report [importance:40 dev:70]
- HarvestBench: Measuring Whether LLM Agents Will Pay to Avoid Killing Animals [importance:35 dev:55]
- The Normalization of Deviance in AI Development [importance:40 dev:40]
- CIT-CAD: Constraint Intent Tree-based CAD Code Generation and Verification [importance:45 dev:70]
- FrogNano: Training a 4B Coding Agent via Online Task Synthesis [importance:60 dev:80]
- From Event Logs to Governed Action: A BlueSky Agenda for Agentic Process Mining [importance:45 dev:70]
- Beyond Coherence: Benchmarking Professional Editing-Technique Execution in Multi-Shot Audio-Video Generation [importance:35 dev:55]
- Valerant: An Automatic Navigable Game Map Generator via Action-Conditioned World Model Exploration [importance:30 dev:60]
- Benchmark Radar: A Living Database and Search Engine for AI Benchmarks and Evaluation [importance:55 dev:80]
- From Explanations to Interventions: Execution-Guided Counterfactual Synthesis in Temporal Graphs [importance:40 dev:65]
- SemVerBench: Benchmarking LLM Comprehension of Version-Constraint Resolution Semantics [importance:50 dev:75]
- A Voice-Interactive Multi-Agent System for Smart Operating Rooms: Architecture Design and Key Technologies [importance:30 dev:60]
- BlueLM-GUI Technical Report: A Real-Device-Centric Flywheel for Self-Improving Mobile GUI Agents [importance:65 dev:80]
- Beyond Generation and Accuracy: Diagnosing and Enhancing Visual Chain-of-Thought for Geometry Problem Solving [importance:45 dev:70]
- K-Bench: A Benchmark for LLM Unlearning in Agentic Deployments [importance:60 dev:80]
- SynGhost: Invisible and Universal Task-agnostic Backdoor Attack via Syntactic Transfer [importance:30 dev:50]
- Zonal RL-RRT: Integrated RL-RRT Path Planning with Collision Probability and Zone Connectivity [importance:25 dev:50]
- CHAI for LLMs: Improving Code-Mixed Translation in Large Language Models through Reinforcement Learning with AI Feedback [importance:35 dev:65]
- From Voice to Value: Leveraging AI to Enhance Spoken Online Reviews on the Go [importance:25 dev:45]
- Balancing Global Quality and Pronoun-Specific Feedback for Context-Aware Machine Translation [importance:30 dev:60]
- EventVL: Understand Event Streams via Multimodal Large Language Model [importance:40 dev:65]
- Towards Unified Approaches in Self-Supervised Event Stream Modeling: Progress and Prospects [importance:35 dev:65]
- LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers [importance:45 dev:70]
- Sanity Checking Causal Representation Learning on a Simple Real-World System [importance:40 dev:60]
- L-Lipschitz Gershgorin ResNet Network [importance:25 dev:60]
- BGM2Pose: Active 3D Human Pose Estimation with Non-Stationary Sounds [importance:25 dev:55]
- Lost-in-the-Middle in Long-Text Generation: Synthetic Dataset, Evaluation Framework, and Mitigation [importance:55 dev:75]
- Generative Learner for Distributional Causal Effects [importance:30 dev:60]
- ModiGen: A Large Language Model-Based Workflow for Multi-Task Modelica Code Generation [importance:40 dev:70]
- WaveHiTS: Wavelet-Enhanced Hierarchical Time Series Modeling for Wind Direction Nowcasting in Eastern Inner Mongolia [importance:25 dev:55]
- Are explainable AI (XAI) evaluation strategies aligned? Comparing subjective, objective, and mathematical evaluation measures using saliency maps [importance:40 dev:65]
- JSolver: Joint Spectrum Estimation and Multi-Material Decomposition from Single-Energy CT Projections [importance:20 dev:50]
- Dissociating performance from compositional feature learning [importance:40 dev:65]
- Unraveling the iterative CHAD [importance:25 dev:60]
- Tensorization is a powerful but underexplored tool for compression and interpretability of neural networks [importance:45 dev:60]
- Discovering Hierarchy-Grounded Domains with Adaptive Granularity for Clinical Domain Generalization [importance:25 dev:50]
- LLM Probability Concentration: How Alignment Shrinks the Generative Horizon [importance:50 dev:55]
- Can Interpretation Predict Behavior on Unseen Data? [importance:45 dev:65]
- Data Security in Large Language Models: Risks, Defense, and Directions [importance:55 dev:65]
- Perceptual Reality Transformer: What Must an Illustration Preserve? [importance:25 dev:40]
- Access Paths for Efficient Ordering with Large Language Models [importance:45 dev:70]
- Exploring the Potential of Diffusion Large Language Models in Code Generation [importance:55 dev:75]
- Tackling GNARLy Problems: Graph Neural Algorithmic Reasoning Reimagined through Reinforcement Learning [importance:40 dev:65]
- Uncertainty-Aware Calibrated Clinical Text Classification with Large Language Models [importance:20 dev:50]
- Concertina: Data-Centric Adaptive Pipeline Parallelism for Efficient Heterogeneous Long-Context LLM Training [importance:60 dev:80]
- IsingFormer: Augmenting Parallel Tempering With Learned Proposals [importance:30 dev:50]
- Generalized Correctness Models: Learning Calibrated and Model-Agnostic Correctness Predictors from Historical Patterns [importance:55 dev:65]
- TOPO-Bench: An Open-Source Topological Mapping Evaluation Framework with Quantifiable Perceptual Aliasing [importance:35 dev:50]
- CapGeo-Bench: Decoupling Visual Perception from Reasoning and Evaluating Geometric Understanding [importance:45 dev:60]
- ParsVoice: A Large-Scale Multi-Speaker Persian Speech Corpus for Text-to-Speech Synthesis [importance:30 dev:60]
- Interpretable Recognition of Cognitive Distortions in Natural Language Texts [importance:20 dev:40]
- Mesh-based Super-resolution of Multiscale Detonation Flows with Graph Transformers [importance:25 dev:55]
- Freeze, Share, Shrink: Rethinking the Action Backbone in Diffusion Policies [importance:40 dev:65]
- Donors and Recipients: On Asymmetric Transfer Across Tasks and Languages with Parameter-Efficient Fine-Tuning [importance:50 dev:70]
- Can We Stop Malicious AI? KILLBENCH: A Benchmark for External AI Kill Switch Feasibility [importance:65 dev:70]
- MedSAM3: Delving into Segment Anything with Medical Concepts [importance:25 dev:55]
- DuoTok: Source-Aware Dual-Track Music Tokenization for Vocal-Accompaniment Generation [importance:35 dev:60]
- Understanding the Effects of Distractors on Reasoning Vision-Language Models [importance:45 dev:65]
- How Semantically Stable Are LLM Refusals? Measuring Confusion in Local Safety Boundaries [importance:55 dev:70]
- Enhancing Large Language Model-Based Systems for End-to-End Circuit Analysis Problem Solving [importance:40 dev:50]
- Focus on What Matters: Fisher-Guided Adaptive Multimodal Fusion for Vulnerability Detection [importance:55 dev:75]
- Large Language Models As Shannon Lossy Compressors Not Solomonoff Induction Estimators: The Singularity Is Not Near Without Symbolic Model Synthesis in Program Space [importance:30 dev:50]
- Confident Rankings with Fewer Items: Adaptive LLM Evaluation with Continuous Scores [importance:45 dev:70]
- Human Values in a Single Sentence: Moral Presence, Hierarchies, and Transformer Ensembles on the Schwartz Continuum [importance:30 dev:55]
- TIDAL: Temporally Interleaved Diffusion and Action Loop for High-Frequency VLA Control [importance:50 dev:75]
- SDUs DAISY: A Benchmark for Danish Culture [importance:25 dev:45]
- AbFlow : End-to-end Paratope-Centric Antibody Design by Interaction Enhanced Flow Matching [importance:25 dev:50]
- Dreaming in Code for Curriculum Learning in Open-Ended Worlds [importance:45 dev:65]
- Learning Human-Like Badminton Skills for Humanoid Robots [importance:30 dev:55]
- Rubrics as an Attack Surface: Stealthy Preference Drift in LLM Judges [importance:60 dev:75]
- You Can Learn Tokenization End-to-End with Reinforcement Learning [importance:55 dev:75]
- Context-Dependent Affordance Reports in Vision-Language Models [importance:40 dev:60]
- Speech Generation Speaker Poisoning: Capability Erasure in Zero-Shot Text-to-Speech [importance:50 dev:65]
- Hindsight-Anchored Policy Optimization: Learning Through Hindsight with Thompson Sampling-Inspired Adaptive Gating [importance:50 dev:70]
- Surprised by Attention: Predictable Query Dynamics for Time Series Anomaly Detection [importance:40 dev:60]
- GP-VM$\times$SMA: Benchmarking General-Purpose Vision Models and Specialized Architectures for 2D Medical Image Segmentation [importance:25 dev:60]
- To See is Not to Master: Teaching LLMs to Use Private Libraries for Code Generation [importance:60 dev:80]
- Measuring and Exploiting Contextual Bias in LLM-Assisted Security Code Review [importance:55 dev:75]
- GT-Space: Enhancing Heterogeneous Collaborative Perception with Ground Truth Feature Space [importance:50 dev:70]
- AI Psychosis: Does Conversational AI Amplify Delusion-Related Language? [importance:35 dev:55]
- Testing the Limits of Truth Directions in LLMs [importance:45 dev:70]
- Can We Still Trace L1 Signals? Investigating the Resilience of Native Language Signals in the LLM Era [importance:35 dev:55]
- SatIR: Scalable High-Recall Constraint-Satisfaction-Based Information Retrieval for Clinical Trials Matching [importance:30 dev:60]
- Efficient Personalization of Generative User Interfaces [importance:40 dev:65]
- Pareto-Optimal Offline Reinforcement Learning via Smooth Tchebycheff Scalarization [importance:55 dev:75]
- DSS: Dynamic Semantic Steering for Robust Concept Erasure in Diffusion Models [importance:55 dev:70]
- An AI Agent Execution Environment to Safeguard User Data [importance:65 dev:80]
- Useless but Safe? Benchmarking Utility Recovery with User Intent Clarification in Multi-Turn Conversations [importance:50 dev:70]
- Putting HUMANS first: Efficient LAM Evaluation with Human Preference Alignment [importance:40 dev:65]
- Human-in-the-Loop Meta Bayesian Optimization for Fusion Energy and Scientific Applications [importance:35 dev:60]
- Adaptive 3D-RoPE: Physics-Aligned Rotary Positional Encoding for Wireless Foundation Models [importance:45 dev:70]
- The Garden of Forking Paths: Threading Narrative Archetype as a Semantic Signal Through Gameplay Planning [importance:35 dev:55]
- Data driven approach for Outdoor Channel Prediction in 5G and Beyond [importance:40 dev:65]
- IntraGuard: Committee-Side Defenses Against Review Outsourcing to Commercial Chatbots [importance:45 dev:60]
- Deep Tech to Space: Space Data Centers and AI Revolution at the Edge [importance:45 dev:70]
- Transcoders Trace Visual Grounding and Hallucinations in Vision-Language Models [importance:50 dev:75]
- Mimir: Large-scale Multilingual Concept Modeling [importance:55 dev:75]
- LoSATok: Low-Dimensional Semantic-Acoustic Tokenizer for Cross-Domain Audio Understanding and Generation [importance:50 dev:75]
- Agentic AI for Gravitational Wave Data Analysis: A Head-to-Head Comparison of Coding Agents Executing a Matched Filter Pipeline on Einstein Telescope Simulated Data [importance:55 dev:75]
- Compute Allocation for Self-Evolving LLMs: From Depth-Breadth to Multi-Armed Bandits [importance:50 dev:75]
- Generating Reports or Repeating Templates? Measuring and Mitigating Template Collapse in 3D CT Report Generation [importance:35 dev:65]
- Emergence of Exploration in Policy Gradient Reinforcement Learning via Retrying [importance:50 dev:75]
- InfoAtlas: A Foundation Model for Zero-Shot Statistical Dependence Estimate [importance:50 dev:75]
- FVSpec: Real-World Property-Based Tests as Lean Challenges [importance:55 dev:80]
- Policy and World Modeling Co-Training for Language Agents [importance:60 dev:80]
- Attention Calibration for Position-Fair Dense Retrieval [importance:45 dev:70]
- Qwen-Image-Flash: Rethinking the Training Recipe for Few-Step Distillation [importance:55 dev:80]
- EasyLens: A Training-Free Plug-and-Play Subtle-Lesion Representation Amplifier for Medical Vision-Language Models [importance:30 dev:65]
- Hearing the Unspoken: Language Model Priors for Acoustic Adversarial Attacks [importance:50 dev:70]
- Cherry-pick Override: LLM Judges Under-use the Non-Directional Verdicts Their Contract Authorizes [importance:45 dev:65]
- FiberTune: Preserving Action-Fiber Visual Residuals in Vision-Language-Action Fine-Tuning [importance:50 dev:75]
- SAILS: Surrogate-based Analysis of Interactions via Local Effect Smooths [importance:50 dev:70]
- How Do Video Foundation Models Encode Intuitive Physics? Probing Across Pretraining Paradigms [importance:45 dev:65]
- Attention-Discounted Adaptive Sampler for Masked Diffusion Language Models [importance:55 dev:80]
- AgentRivet: an automated system for producing Rivet routines from journal publications [importance:30 dev:55]
- Mask, Sample, Revise: A Revisable CTMC Inference Stack for Guided Discrete Flow Matching Text-to-Speech [importance:50 dev:80]
- IUU+DB: Tracking Illegal, Unreported, and Unregulated Fishing, Seafood Fraud, and Labor Abuse through LLM-driven Information Extraction [importance:45 dev:70]
- Robusto-2: Benchmarking Humans & VLMs for Autonomous Driving in Lima & New York City [importance:50 dev:75]
- UniRank: Unified Rank Allocation for Low-Rank LLM Compression [importance:55 dev:80]
- The Language-Energy Divide: Measuring Energy Costs of Multilingual LLM Inference [importance:50 dev:75]
- DeepDiscovery: A Location-Inference Framework for Task-Level Repository Understanding [importance:60 dev:80]
- Weave of Formal Thought [importance:60 dev:80]
- Interpretable Inverse Design of Metal-Organic Frameworks with Large Language Model Agents [importance:35 dev:65]
- DART-VLN: Test-Time Memory Decay and Anti-Loop Regularization for Discrete Vision-Language Navigation [importance:45 dev:70]
- Mapping Text to Multiplex Graph: Prompt Compression as L\'evy Walk-Guided Graph Pruning [importance:50 dev:75]
- PEEK: Predictive Queue-Informed KV Cache Management for LLM Serving [importance:65 dev:85]
- DualView: Preventing Indirect Prompt Injection in Personal AI Agents [importance:65 dev:85]
- Energy Accuracy Is Not Enough: A Structure-Aware Benchmark and Evaluation Protocol for Quantum Architecture Search [importance:30 dev:60]
- REDDIT: Forgetting-Resistant Correction of Timestamp Drift in ASR via Replay-Based Distribution Editing [importance:50 dev:75]
- LEXIC: Lightweight On-Device Decoding of Reading Comprehension from Eye Movements [importance:30 dev:55]
- SMetric: Rethink LLM Scheduling for Serving Agents with Balanced Session-centric Scheduling [importance:65 dev:85]
- TypiCore: A Hybrid Active Query Strategy for Class-Incremental Learning on Time Series [importance:40 dev:70]
- Adaptive Adversaries: A Multi-Turn, Multi-LLM Benchmark for LLM Agent Security [importance:60 dev:80]
- SechKAN: Kolmogorov-Arnold Networks with Hyperbolic Secant Functions [importance:45 dev:70]
- Freezing the Physiological Encoder: Explanation Stability Under Bounded Updates of an ICU Model [importance:30 dev:10]
- REGEN: Replay-recycling for Expert-to-Generalist distillation with Offline Reinforcement Learning [importance:45 dev:70]
- X-Stage: Modeling Post-Issue Backpressure in GPU Communication--Computation Fusion [importance:45 dev:70]
- MulRobBench: A Decision-Level Benchmark for Safe and Security-Policy-Compliant Multimodal UAV Agents [importance:40 dev:65]
- Looping Is Not Reliability: State-Bound Evidence and Typed Revision Contracts for Agentic Code Repair [importance:60 dev:75]
- Learning from 53.6K Real-World Developer Edits of AI-Generated Code [importance:65 dev:80]
- FairFund-Bench: Evaluating Distributive Bias in LLM Resource Allocation [importance:55 dev:70]
- DreamQAS: Learning a Decision-Useful World Model for VQE-Efficient Quantum Architecture Search [importance:40 dev:60]
- FriendBench: Benchmarking Dyadic Familiarity Inference in Humans and Multimodal Large Language Models [importance:45 dev:70]
- Can AI Agents Simulate A/B Test Outcomes? A Validation Framework for Agentic Experimentation [importance:55 dev:75]
- MameLoshnLM: Yiddish Language Model and Evaluation Benchmark [importance:25 dev:70]
- Hidden Gauge Controls Feature Specialization in ReLU Networks [importance:40 dev:60]
- Ready Cohorts: Bounding GPU Opportunity and Avoiding Host Round Trips in LLM-Agent Control [importance:60 dev:80]
- FluctlightDB: A Memory Model of Data for AI Agents [importance:65 dev:80]
- KREL: Automatic Medical Coding via Knowledge-Guided Reasoning over Clinical Evidence with LLMs [importance:30 dev:50]
- What's the Catch? Evaluating Temporal Consistency in Vision-Language Models [importance:45 dev:70]
- A Few Pages of Markdown: Committed AI Configuration and Lower Quality Cost after Coding-Agent Adoption [importance:70 dev:85]
- CRAMER: Control via Request-Aware Masking for Editing Recommenders [importance:35 dev:65]
- Syn2Logic: End-to-End Neuromorphic Design Automation [importance:35 dev:65]
- When the Canonical Completion Is Wrong: Formalizing and Measuring the Jump in Large Language Models [importance:50 dev:75]
- Simultaneous Envy and Equitability Guarantees [importance:20 dev:40]
- Persona-Execution Separation: An Architecture Pattern for Evolving LLM Agents under Execution Audit [importance:70 dev:85]
- A rigor-matched audit of periodic-step layer skipping for efficient llm inference: conflayers versus swift, with a supplemental analysis of trained routing alternatives [importance:55 dev:80]
- Untangling the Mechanisms of Misleading Context in Medical Question Answering [importance:25 dev:50]
- Adaptation Interfaces for In-Context Tabular Foundation Models in Time-to-Event Prediction [importance:30 dev:50]
- Skynet: Workflow-Level Anomaly Detection for Agentic AI via Semantic and Structural Modeling [importance:65 dev:80]
- Constrained Online Learning with Noisy Constraint Values [importance:35 dev:65]
- Proprioception-Anchored Cross-Modal Pretraining for Zero-Shot Sim-to-Real Contact-Rich Assembly [importance:45 dev:70]
- Multimodal Duplex Interaction Agent [importance:55 dev:80]
- ThinkPrior: Zero-Rollout Difficulty Priors for Cold-Start Prompt Selection in RLVR [importance:50 dev:75]
- No Free Checker: A Survey of Verifiers for Robot Policies [importance:40 dev:70]
- ActSafeGuard: Differentiable and Training-Aligned Constraint Enforcement for Flow-Matching Policies [importance:50 dev:75]
- Agentic TCAD Calibration Workflow for Oxide Semiconductor Transistors [importance:45 dev:75]
- Evaluating Context Segmentation in Locally Deployable SLMs for Cybersecurity CTF Tasks [importance:50 dev:75]
- Causal neural set filtering for online multi-target tracking [importance:45 dev:75]
- Managing Action Preconditions in Neuro-Symbolic RL: Three Placement Strategies for Embodied Agents [importance:55 dev:80]
- OmniHarness: Harnessing Generalizable Visual Generation via Symbolic Policy Learning [importance:50 dev:75]
- Driver Behavior Estimation at Signalized Intersections Using a Physics-Constrained Decision-Conditioned Autoregressive Transformer [importance:40 dev:65]
- HintMiner: Automatic Question Hints Mining From Q&A Web Posts with Language Model via Self-Supervised Learning [importance:35 dev:70]
- POSPAN: Position-Constrained Span Masking for Language Model Pre-training [importance:40 dev:75]
- Signed p-adic Residual Encodings of Finite-Domain All-Different Systems with a Sudoku Case Study [importance:35 dev:70]
- You Don't Need To Train: Agentic Heuristic Learning Studio for Executable Human Activity Recognition [importance:20 dev:40]
- A panoramic aerodynamic performance prediction method for turbomachinery cascades using transformer-enhanced neural operator [importance:45 dev:70]
- A Dynamic Aggregation Strategy Enhanced Efficient Global Optimization Algorithm for Solving High-Dimensional Turbomachinery Design Problems [importance:40 dev:70]
- Beyond Distribution Matching: Semantics-Consistent Tabular Diffusion with Weak Semantic Priors [importance:45 dev:70]
- Schema-Adaptive Action-Conditioned JEPA for Cross-Machine CNC Transfer under Partial Sensor Overlap [importance:50 dev:75]
- Pseudo-Label Augmentation for Affect Sensing in Small Collaborative Groups [importance:40 dev:65]
- Distilling Foundation Models for Agentic What-If Reasoning:Cost, Latency, and Governance in a Hybrid LLM+SLM Architecture [importance:60 dev:80]
- Evaluating Open-Weight E-Commerce Agents with Environment-Grounded Verification [importance:55 dev:75]
- SWB-DM: A Calibrated Sliced-Wasserstein-Barycenter Aggregator with Delayed-Momentum Caching for Byzantine-Robust Federated Learning under Partial Participation [importance:45 dev:70]
- A Decision-Support Audit Protocol for Supervision Drift in Proxy-Labeled Credit-Risk Prediction [importance:50 dev:75]
- LLMs as Master Forgers: Generating Synthetic Time Series Data for Manufacturing [importance:45 dev:70]
- LLM Inference in a Flash! [importance:65 dev:85]
- Skeletal Prototypes on Iterative Nerve Expansions [importance:40 dev:65]
- Z-Loss Backward Geometry in Dense Output Heads and Sparse Routers [importance:50 dev:75]
- Anatomy of Associative Recall in Fixed-State Recurrences: A Matched-State Decomposition, an Interference Wall, and a Curriculum That Breaks It [importance:55 dev:80]
- Decoy Direction Optimization: A Post-Hoc Defense Against LLM Abliteration [importance:60 dev:80]
- How I learned to stop worrying and love StopGrads: Stationarity, Convergence, and a case study on Flow Map Learning [importance:50 dev:75]
- Test-Time Unlearning via Sparse Autoencoder [importance:55 dev:80]
- Efficient Reasoning Distillation: Small Video-Language Models via Synthetic CoT and Difficulty-Aware Fine-Tuning [importance:50 dev:75]
- The record is part of the task: matched-record evaluation of text classifiers across maintenance, safety and recall reporting [importance:40 dev:70]
- Scaling Laws for Physics-Aware ACOPF Surrogate Learning [importance:45 dev:70]
- Differentially Private Semantic Plans for Aggregate Insight Generation [importance:45 dev:70]
- Drift Field Net: Learning Ocean Lagrangian advection fields from in-situ and satellite observations [importance:35 dev:60]
- Agentic Search Spaces for Tabular Machine Learning [importance:60 dev:80]
- Robust Fault Detection in Mechanical Multimodal Time Series via Self-Supervised Cross-Modal Reconstruction [importance:45 dev:70]
- Generative models for simulation based filtering: Formulations and Empirical Comparisons [importance:45 dev:70]
- Channel-Informed Neural Network for Physical Layer Key Generation [importance:45 dev:75]
- Multi-Label Proportion Learning for Sea-Ice Type Prediction [importance:40 dev:65]
- Federated stochastic bilevel optimization with fully first-order gradients [importance:45 dev:70]
- Autonomous Droplet Navigation via Model-Based Reinforcement Learning [importance:45 dev:70]
- Certified Uncertainty Propagation in One-Shot Federated Bayesian Models via Posterior Event Transport [importance:50 dev:75]
- Bounded Adjustment with Reliability-Guided Embedding for Imbalanced Learning with Noisy Labels [importance:45 dev:70]
- Attention Mean Fields Predict Average Representation Dynamics and Reveal Context-Specific Computation [importance:55 dev:80]
- How Good Are Time-Series Foundation Models for Pedestrian Crowd Count Forecasting? A Cross-Dataset Comparative Study [importance:45 dev:70]
- Interpreting and Steering LLM Agents for Social Simulations [importance:55 dev:75]
- Adaptive Bayesian Partner Selection for Federated Clinical Centers [importance:50 dev:75]
- OPD-Aha: From Linguistic Momentum to Visual Reflection in Multimodal On-Policy Distillation [importance:55 dev:75]
- Online Gradient Computation for Warping Gaussian Process Transformations [importance:45 dev:70]
- Decoder Design Matters for ECG Delineation [importance:45 dev:70]
- High-Performance Tensor Formulation of the Viterbi Algorithm for Hidden Semi-Markov Models [importance:50 dev:75]
- FlowATC: Aircraft Trajectory Prediction via Flow Matching [importance:50 dev:75]
- What Does Layer-Importance Reveal About Transformers and State-Space Models? [importance:55 dev:80]
- On the Importance of Gating: Memorization vs. In-Context Learning in State Space Models [importance:60 dev:80]
- AsyncCouple-Flow: Asynchronous Cross-Modal Coupling and Flow Matching for Spatio-Temporal Forecasting [importance:50 dev:75]
- Recovering Physical Parameters from Fragmented Observations via Exact Distributed Spline Merging [importance:40 dev:65]
- A Weighted Kernel Method for Approximation that Adapts to Learned Multivariable Structure [importance:45 dev:70]
- Divergence Timing and Cumulative Disagreement under KV-Cache Eviction [importance:55 dev:80]
- Stable by Construction: Variational Latent Markov Operators for Long-Horizon PDE Prediction [importance:50 dev:75]
- GrowMTP: Can RL Grow Its Own Draft Head? [importance:60 dev:80]
- Right Direction, Wrong Step: Geometric Analysis of Finite-Step Failure in Looped Transformers [importance:55 dev:80]
- Continuous-Time Machine Learning: A Unified Mathematical Perspective [importance:50 dev:75]
- A Systematic Evaluation of Machine Learning Methods for Fault Detection and Line Identification in Electrical Power Grids [importance:45 dev:70]
- TAME: Token Attribution and Masking for Emergent misalignment [importance:60 dev:80]
- Noise2Noise Revisited: Training Pair Distributions Dominate Loss Choice in Self-Supervised Denoising [importance:45 dev:75]
- SOTER: A Generative Time-Series Foundation Model for Wearable Human Physiological Signals [importance:50 dev:75]
- Geometry of learning dynamics: Gradient descent versus natural gradient on the ridge of optimization [importance:50 dev:75]
- ImpossibleRubrics: Stress-Testing Generated Rubrics as Reward Signals [importance:55 dev:75]
- LCAP: Population-Informed Latent Chip Adaptation from Few Output Probes for Photonic Neural Networks [importance:50 dev:75]
- Adapting to Decision-Relevant Non-Stationarity in Decentralized Heterogeneous Bandits [importance:45 dev:70]
- Information Geometric Self-Organization at the Edge of Stability in High-Capacity Kernel Associative Memories [importance:20 dev:50]
- Can Deep Learning Achieve Cross-Physics Mapping? [importance:25 dev:45]
- HyCoSeq: Contextual Hyperbolic Representation Learning for Genomic Sequences [importance:20 dev:45]
- Verbalizing Subliminal Learning Effects Using Text Optimization [importance:35 dev:60]
- Repurposing Deep Limit Order Book Forecasting for Scenario-Conditioned Market Impact Modeling [importance:20 dev:50]
- When Confidence Signals Disagree: Local and Global Confidence in Autoregressive Language Models [importance:45 dev:65]
- Beyond Token-Local Imitation: Reward-Compatible Temporal Credit Assignment for On-Policy Distillation [importance:50 dev:70]
- Structural Negative Transfer in Federated Graph Neural Networks: Diagnosis, Causal Investigation, and the Limits of Divergence-Aware Mitigation [importance:30 dev:65]
- CLARE: Scalable Class-Incremental Continual Learning via a Sparsity-Based Framework [importance:35 dev:65]
- Distributed JEPA: A Self-Supervised Framework for Energy Forecasting [importance:30 dev:60]
- Learning Options for Compositional Motor Control with Adapter Banks [importance:30 dev:60]
- Repurposing Unified Topological Signatures for Graph Representation Learning [importance:35 dev:65]
- High-Fidelity Digital Twin Data Models by Randomized Dynamic Mode Decomposition and Deep Learning with Applications in Fluid Dynamics [importance:35 dev:55]
- Neural Field Ensembles for Aerodynamic Surface Prediction: Winning Solution to the ONERA CRM Wall Distribution 2025 Challenge [importance:35 dev:55]
- A unified framework for global and local interpretability using adaptive derivative-ordered random explanation [importance:45 dev:70]
- IRENE: A Convolutional GRU Ensemble Model for Radar Precipitation Nowcasting over Italy [importance:25 dev:55]
- LoopSpec: Pipelined Self-Speculative Decoding for Looped Transformers [importance:55 dev:75]
- MyoFlow: Anchor-Tied Rectified Flow for HD-sEMG Gesture Recognition Across Sessions and Subjects [importance:25 dev:55]
- Memorisation bias in medical AI [importance:45 dev:60]
- Easy to Catch a Liar, Hard to Clear an Honest One: Language Models Diagnosing a Corrupted Reward Channel from a Verified Record [importance:45 dev:70]
- Personalized Federated Learning through Global Knowledge Distillation and Local Head Adaptation [importance:35 dev:65]
- Same Flow, Different Paths: Variance Reduction in Flow Matching [importance:40 dev:70]
- Hybrid Variational Quantum Circuits for Multivariate Regression and High-Dimensional Data Reconstruction [importance:30 dev:65]
- Large Language Models Develop Belief State Geometry In-Context [importance:50 dev:70]
- OPEN-1B: A Fully Auditable Training Run [importance:55 dev:75]
- Bridging the Confidence Gap: Temperature Scaling for Calibrating Test-Time Prompt Tuning [importance:40 dev:70]
- Knowledge as Orbit: Finite Collections as Phases of an Exactly Periodic Latent Generator [importance:35 dev:70]
- Learning-Guided Planning in Large Dynamic Action Spaces: Budgeted Tree Search for One-to-Many Mobile Charging [importance:30 dev:65]
- Reduced-Space Multi-Fidelity Bayesian Optimization of Process Simulation Models [importance:30 dev:60]
- Coupled Calibration and Learning: Mitigating Teacher Bias in LLM Distillation without Target-Domain Reward Feedback [importance:50 dev:75]
- FreqSpaNet: Frequency and Spatial Learning of SFPF for Physical Layer Hardware Integrity Detection [importance:30 dev:60]
- ENCP: Episode-Normalized Conformal Prediction for Vision-and-Language Navigation [importance:30 dev:65]
- Nonsmooth Optimization via Orthogonalized Momentum [importance:40 dev:75]
- Few-Shot Degradation Is Not What It Seems: Behavioral Evidence, Representation Analysis, and a Random-Text Control Across 12 Models, 2 Tasks, and 2 Architectures [importance:45 dev:70]
- Single Document Extractive Summarization using Domination in Hypergraph [importance:30 dev:65]
- Latent Undertow: How Ordinary Typos Break Probes [importance:45 dev:75]
- Crash Narrative-Guided Countermeasure Recommendation Using Large Language Models: A Retrieval-Augmented Generation Framework for Intersection Safety [importance:35 dev:60]
- Measuring AI harms with multidimensional Lorenz Zonoids [importance:45 dev:65]
- EMODY Flow: Emotion-Aware Audio-Driven Full-Body Motion Generation [importance:30 dev:60]
- ViCo: Visual-oriented Coding with Self-Reflection for Chart Replication [importance:40 dev:75]
- Are We Grading Properly? Understanding Failure Modes in Medical Benchmarks [importance:40 dev:70]
- 3D Field Data Reduction with Adaptive Sample-Based Gaussian-Encoded Reconstruction [importance:30 dev:65]
- Molecular representation shapes the balance between target fidelity and exploration in flow based polymer generation [importance:30 dev:60]
- A deep dictionary network-based foundation model for ultra-low-dose CT denoising [importance:35 dev:60]
- Towards Scalable RLVR: Multimodal Instruction Following Data Synthesis and Distillation [importance:50 dev:75]
- Digital Persuasion: Understanding the Impact of Online Influencers on Public Opinion [importance:15 dev:30]
- Predicting Social Media Engagement using Machine Learning [importance:25 dev:55]
- Is INT8 Portable? A Cross-Platform Measurement Study of Quantized Inference on Embedded and Automotive Accelerators [importance:50 dev:75]
- Safe Error Correction for Language Models: Frozen-Base Adjustment with Capability Preservation [importance:45 dev:75]
- Computer-assisted global regularity across nonlinear families of three-dimensional periodic Navier-Stokes flows [importance:30 dev:50]
- GPEvac: GNN-Based PPO for Adaptive Evacuation Routing During Shooting Events [importance:35 dev:70]
- Permutation-Based Stegomalware in Large Language Models: Threats and Countermeasures [importance:50 dev:75]
- A Sentinel-2 benchmark dataset for deep-learning active-fire segmentation across 25 California wildfires [importance:30 dev:65]
- Improving Reduced-Order Rotating Detonation Engine Models with Data Assimilation and Machine Learning [importance:25 dev:60]
- Copula Adapted Directed Acyclic Graph for Cluster Representation of Biomedical Data [importance:30 dev:65]
- The AI-Enabled Scientific Frontier [importance:55 dev:60]
- Compute-Optimal Pretrain--Fine-tune in Ridge Gradient Descent [importance:50 dev:75]
- Towards Surrogate Based Dequantization of Quantum Reinforcement Learning [importance:25 dev:70]
- Semantic-Aware Neural Video Codec for Error-Resilient Low-Latency Transmission [importance:40 dev:70]
- Symmetric solution of the Bellman optimality equation for repeated harmony game [importance:25 dev:70]
- Nationally Consistent, Locally Incomplete: A Bayesian Remote-Sensing Audit of Rooftop Photovoltaic Registries [importance:25 dev:55]
- BLINDSPOT: A Benchmark for Safety and Refusal Calibration in Long-Horizon Tool-Using Agents [importance:55 dev:75]
- Sequence Recognition in Bharatnatyam dance [importance:15 dev:55]
- FairLint-DL: An IDE-Native Tool for Fairness Debugging of Deep Learning Software [importance:50 dev:75]
- Cross-Anatomy Transfer Versus Sparse Interpolation in Digital-Twin-Oriented Aortic Fluid-Structure Interaction Surrogates [importance:30 dev:65]
- Breaking the 1.58-bit Barrier for Ternary LLMs [importance:55 dev:75]
- StalePO: Anchored Token-Level Preference Optimization using Legacy Post-Edits in Machine Translation [importance:40 dev:70]
- EBL: Efficient Broad Learning for Distributed Adaptive Harmonic Analysis [importance:30 dev:60]
- Mini-batch Sampling Strategies for Long-Tailed Image Classification: An Empirical Study on CIFAR-100-LT [importance:35 dev:70]
- Fast-Convergent Meta-RL via Gradient-Clustered BS Sampling for Edge Caching [importance:35 dev:70]
- Implementing a White-Box Undetectable Backdoor for Random Fourier Features [importance:45 dev:75]
- Physics Informed Random Feature Neural Networks for Solving PDEs [importance:40 dev:70]
- Balancing Trial and Reorder: A Hybrid Sequential Transformer-GBDT Ranker for On-Demand Delivery [importance:40 dev:70]
- On the Expressive Power of Implicit Line-Graph Higher-Order Weisfeiler--Leman [importance:35 dev:70]
- Learned Look-Ahead Splitting Rule for CART [importance:35 dev:70]
- The Neverwhere Visual Parkour Benchmark Suite [importance:35 dev:65]
- Decentralized Gossip Learning and Federated Averaging for Histopathology Image Classification [importance:35 dev:70]
- Early-Bird Decoding: Accelerating Diffusion LLMs with Learnable Block Sizes and Parallel Sampling [importance:50 dev:75]
- Not All Relations Are Equal: Relation-Balanced and Calibrated Graph Learning for Provenance-Based Intrusion Detection [importance:45 dev:70]
- A multimodal large language model for evidence-based autism spectrum disorder screening [importance:40 dev:65]
- Certified Inference and Training for Deep Equilibrium Networks: A Continuation Framework with Polynomial Complexity Guarantees [importance:45 dev:75]
- Skill-based Agentic Evaluation for Real-time Data Science Tasks [importance:50 dev:75]
- From Manual Construction to AI-Driven Scenario Emergence: Rethinking Catastrophe Risk Modeling [importance:40 dev:65]
- AURA: Agentic Diagnosis and Refinement for Production Recommender Systems at Scale [importance:50 dev:75]
- Can Knowledge Transfer Parameters Be Learned? LePoKet for Efficient Robotic Vision [importance:35 dev:70]
- Weave: Learning Whole-Body Dexterous Loco-Manipulation from Human-Object Interactions [importance:40 dev:70]
- Seeing What Matters: Visual Cue Guided Video Planning for Generalizable Robot Navigation [importance:35 dev:70]
- Unified Heterogeneous Graph Neural Network solver for Power Flow, Optimal Power Flow and State Estimation [importance:40 dev:70]
- Carry-Through Checksum: A Lightweight Fault-Detection for CNN Inference at the Edge [importance:45 dev:75]
- The Latent That Never Was: A Forensic Re-run of the CVAE Ablation in Action Chunking Transformer [importance:35 dev:70]
- Constant Swap Regret in General-Sum Games via Optimistic Transition Matrices [importance:30 dev:70]
- Time-warping estimation via stationarity-based learning of the de-warped signal [importance:25 dev:65]
- On the disintegration of the stochastic majority vote: From PAC-Bayesian bounds to a self-bounding algorithm [importance:35 dev:70]
- Measuring Annotation Efficiency for Handwritten Devanagari Recognition: Sample-Complexity Curves for Four Pretraining Regimes [importance:25 dev:65]
- TEMPO: Learning Temporal Context for Dynamic Robot Manipulation [importance:45 dev:75]
- NeuroTS-Net: Multi-Class Semantic Segmentation of Pediatric Brain Tumors in Multi-Modal MRI [importance:35 dev:65]
- OptiPrime: Optimizing Private Inference through Protocol-Hardware Co-design [importance:50 dev:75]
- Multi-Agent Learning with Cooperation-Driven Optimization Dynamics [importance:35 dev:70]
- Causal Discovery via Transformed Low-Rank Quantile Surfaces [importance:35 dev:70]
- MedPCFM-TED: One-Step Point Cloud Flow Matching for Implant Generation via Teacher-Guided Endpoint Distillation [importance:35 dev:70]
- HUMAID-NER: A Disaster Tweet Dataset for Joint Named Entity Recognition and Event Classification via Uncertainty-Weighted Multitask Learning [importance:15 dev:25]
- Splitting the Difference: Interpretable Causal Forests for Treatment Effect Heterogeneity and Bias [importance:20 dev:45]
- Beyond Measurement Metrics: A Human-Centered Framework for Semantic Validation of Network Traffic Classification [importance:28 dev:55]
- Near-Optimal Nonconvex Matrix Completion [importance:12 dev:38]
- Neuro-Symbolic Hierarchical Intention Anticipation in Human Behavior [importance:38 dev:58]
- Bio-Inspired Palette Evolution in Indirectly Encoded Substrates: Timescale Compatibility Shapes Activation Function Discovery [importance:25 dev:55]
- Optimization over covariance matrices with a parameterized metric [importance:14 dev:42]
- Intrinsic Robot Rewarding: Reusing VLA Representations for Autonomous Evaluation and Policy Improvement [importance:45 dev:65]
- From Foundation Embeddings to Cropland Maps: Label Efficiency, Temporal Transferability and Independent Human Validation [importance:35 dev:55]
- Continual Learning for Traversability Prediction with Uncertainty-Aware Adaptation [importance:40 dev:60]
- Kernel-Based Metrics Learning for Uncertain Opponent Vehicle Trajectory Prediction in Autonomous Racing [importance:33 dev:60]
- ResLRP: The Role of Residual Cancellation in Attribution Instability in Vision Transformers [importance:40 dev:65]
- Cross-Domain Inference for Human Localization: Applying Wi-Fi RSSI Data to CSI-Trained Models [importance:28 dev:42]
- Conformal Policy Learning with Distribution-Free Safety Guarantees [importance:40 dev:65]
- Goal-oriented probabilistic forecasting for dynamic PRB allocation in 5G networks [importance:32 dev:55]
- Quantum-Inspired Trainable and Parameter-Efficient Tensor Networks for Image Inpainting [importance:28 dev:60]
- Type-IV Code Clone Detection via Layer-Wise Non-Contrastive Representation Learning [importance:40 dev:75]
- Tables Decoded: DELTA for Structure, TARQA for Understanding [importance:45 dev:65]
- Bias-Induced Crossover in Absolute Capacity of Dense Associative Memory [importance:18 dev:42]
- Bridging the Gap Between Homogeneous and Heterogeneous Asynchronous Optimization Is Surprisingly Difficult [importance:22 dev:55]
- Robust Recurrent Reinforcement Learning under Evolving Hidden Disturbances with Application to Rover Wheel Slip [importance:38 dev:68]
- Attention is All You Need Until You Need Retention [importance:50 dev:75]
- Explainable Graph-theoretical Machine Learning with Application to Alzheimer's Disease Prediction [importance:32 dev:55]
- BenSParX: A Robust Explainable Machine Learning Framework for Parkinson's Disease Detection from Bengali Conversational Speech [importance:32 dev:45]
- When majority rules, minority loses: bias amplification of gradient descent [importance:45 dev:70]
- Observational Multiplicity [importance:32 dev:65]
- GraphIFE: Rethinking Graph Imbalance Node Classification via Invariant Learning [importance:38 dev:68]
- GeoCrossBench: Cross-Band Generalization for Remote Sensing [importance:38 dev:65]
- Dual Randomized Smoothing: Beyond Global Noise Variance [importance:38 dev:68]
- Training Energy-Based Models with Non-MCMC Samplers and Efficient Temperature Estimation [importance:28 dev:65]
- Formalized Hopfield Networks and Boltzmann Machines [importance:28 dev:75]
- Collaborative Optimization of Multiclass Imbalanced Learning: Density-Aware and Region-Guided Boosting [importance:35 dev:65]
- Window-Diffusion: Accelerating Diffusion Language Model Inference with Windowed Token Pruning and Caching [importance:50 dev:80]
- Robustness as an Emergent Property of Task Performance [importance:40 dev:68]
- Hybrid Feedback-Guided Optimal Learning for Wireless Interactive Panoramic Scene Delivery [importance:32 dev:55]
- PRISM: Parallel Residual Iterative Sequence Model [importance:55 dev:80]
- Partial recovery of meter-scale surface weather [importance:32 dev:55]
- Strategic Advice in the Age of Personal AI [importance:32 dev:55]
- Routing Absorption in Sparse Attention: Why Random Gates Are Hard to Beat [importance:40 dev:75]
- Learning efficient representations of complex constraints for scalable optimization [importance:38 dev:68]
- Deep Invertible Autoencoders for Dimensionality Reduction of Dynamical Systems [importance:32 dev:65]
- Thinking Deeper, Not Longer: Memory-Efficient Test-Time Reasoning with Depth-Recurrent Transformers for Compositional Generalization [importance:50 dev:80]
- A Spectral Decomposition Framework for Multiscale Nonlinear Dimensionality Reduction [importance:28 dev:65]
- LLM-Guided Dynamic Action Spaces for Synthesizable Molecular Optimization [importance:42 dev:65]
- EviDep: Uncertainty-Aware Multimodal Depression Estimation via Disentangled Evidential Learning [importance:32 dev:55]
- Perturbation Sensitivity of Maximum-Likelihood Pairwise Ranking in Computational Decision Systems [importance:22 dev:60]
- GRAFT-ATHENA: Self-Improving Agentic Teams for Autonomous Discovery and Evolutionary Numerical Algorithms [importance:60 dev:80]
- Structure-Aware Masking for Protein Representation Learning [importance:42 dev:70]
- Benchmarking Machine Learning Architectures for Antimicrobial Stewardship in Pediatric ICUs [importance:32 dev:55]
- BASIS: Batchwise Advantage Estimation from Single-Rollout Information Sharing for LLM Reasoning [importance:55 dev:80]
- Learning aligned EEG representations with subject-specific encoders [importance:32 dev:60]
- Tail-Shape Estimation in LLM Evaluation Is Fragile: A Protocol for Diagnosing False Positives [importance:45 dev:75]
- Amortized Probabilistic Retrieval of Atmospheric CO2 from OCO-2 Spectra Using Deep Learning with Laplace Approximations and Normalizing Flows [importance:32 dev:55]
- The Scissors Effect: When Resize-Based Input Diversity Helps or Hurts Transfer Attacks [importance:32 dev:68]
- Data-Driven Soft Labeling Scales DNA Read Classification to Whole-Body Cell-Type Deconvolution [importance:32 dev:60]
- The Orthogonalized Read Is a Removable Training Scaffold for Recurrent Memory [importance:28 dev:68]
- Neural Operator Learning for Collision-Aware Trajectory Planning of Spacecraft Swarms [importance:38 dev:65]
- Task- and dataset-specific information in protein language models [importance:40 dev:68]
- Activation-Weighted Seeded Residual Coding for Low-Bit LLM Weight Repair [importance:48 dev:78]
- M-Fibration Theory with Applications to Weighted Graphs [importance:10 dev:32]
- ICON Decomposition: Auditing deep neural networks for shortcuts by decomposing layer-wise representations using concepts [importance:45 dev:75]
- Hessian-based molecular conformation augmentation for a scalable and efficient strategy of machine learning interatomic potentials [importance:38 dev:65]
- CoER: Defending against Adaptive Indirect Prompt Injection via Adversarial Co-Evolution and Refinement [importance:55 dev:80]
- Adaptive Anisotropic Attention for Axis-Structured Signals [importance:38 dev:65]
- BRACE: Anchored Bellman-Residual Correction for Stale Critics in Asynchronous RL [importance:50 dev:80]
- Forward-Free LLM Depth Pruning via Weight Redundancy [importance:55 dev:80]
- Meta-LinEXP3: Online-within-Online Learning for Adversarial Linear Contextual Bandits [importance:32 dev:68]
- The Last AI Built by Humans: Toward Genuine Recursive Self-Improvement [importance:75 dev:80]
- From Protocols to Evidence: Bounded Claims for AI in Service of the Common Good [importance:60 dev:75]
- Algorithmic Information Dynamics of Learning: A Certified, Differentiable Complexity Controller for Grokking [importance:40 dev:75]
- The Token Before the Value Is the Key: How Hybrid Architectures Organize Induction Circuits [importance:42 dev:78]
- Discrete Beckmann Transport Models for One-Step Language Modeling and Reasoning [importance:52 dev:80]
- CBW: Towards Dataset Ownership Verification for Speaker Verification via Clustering-based Backdoor Watermarking [importance:42 dev:65]
- R3: Robust Rubric-Agnostic Reward Models [importance:55 dev:80]
- Script Fragmentation and Format: What Drives the English-Bengali Performance Gap in Open LLMs? [importance:48 dev:75]
- Neural Stochastic Differential Equations on Compact State Spaces: Theory, Methods, and Application to Suicide Risk Modeling [importance:32 dev:65]
- Generating Individual Travel Diaries Using Large Language Models Informed by Census and Land-Use Data [importance:32 dev:55]
- Risk-Calibrated Bayesian Streaming Intrusion Detection with SRE-Aligned Decisions [importance:42 dev:68]
- TARC: Time-Adaptive Robotic Control [importance:42 dev:68]
- Shuttling Compiler for Trapped-Ion Quantum Computers Based on Fine-Tuned Large Language Models [importance:55 dev:80]
- Nonnegative matrix factorizations and related compositional models: Equivalence, identifiability, and an application on the grain-size analysis of sediments [importance:22 dev:55]
- AllShowers: One model for all calorimeter showers [importance:42 dev:68]
- Meta-Learning-Assisted Constraint Relaxation for Constrained Black-Box Optimization [importance:38 dev:68]
- Differential privacy representation geometry for medical image analysis [importance:45 dev:75]
- Equivalence of approximation by networks of single- and multi-spike neurons [importance:28 dev:65]
- CodecSight: Leveraging Video Codec Signals for Efficient Streaming VLM Inference [importance:55 dev:80]
- RAM-H1200: A Unified Evaluation and Dataset on Hand Radiographs for Rheumatoid Arthritis [importance:32 dev:55]
- Universal Feature Selection with Noisy Observations and Weak Symmetry Conditions [importance:25 dev:60]
- $\mathcal{O}(n)$ alternative to Quantum Fourier Transform with efficient neural net classical post-processing [importance:42 dev:80]
- Subject-Specific Analysis of Self-Initiated Attention Shifts from EEG with Controlled Internal and External Attention Conditions [importance:25 dev:55]
- Stream Assembly Is an Uncontrolled Treatment in Streaming Intrusion-Detection Benchmarks [importance:32 dev:68]
- HumanEgo: Zero-Shot Robot Learning from Minutes of Human Egocentric Videos [importance:55 dev:75]
- Does Continued Pretraining on a Learner Corpus Improve Automated Essay Scoring on English Proficiency Tests? Evidence from EFCAMDAT [importance:38 dev:68]
- Deep-learning-based low-energy trigger algorithms for the Hyper-Kamiokande experiment [importance:42 dev:68]
- Do LLMs Make Neural Distinguishers Wise? [importance:38 dev:75]
- Shielded Analysis: Certification and Characterization of Defensibility in Systems under Adversarial Interaction [importance:50 dev:80]
- Missing Data Imputation under Manifold Hypothesis [importance:32 dev:65]
- Know Your Agent: Reconnaissance-Driven Pentesting of AI Agents [importance:60 dev:85]
- CausalSmith: A Formally Grounded, Self-Improving Agentic Framework for Automated Research in Causal Inference [importance:65 dev:85]
- Protecting patient privacy in clinical foundation models: Technical and legal perspectives [importance:55 dev:75]
- HoloAegis: Frozen Representation, Topological Inference --- Minimally Parametric Safety Manifolds and Their Capability Boundaries for LLM Guardrails [importance:45 dev:65]
- LaGSplat: Inferring Physics-Governed Interactive Simulation from Monocular Video Using Latent Lagrangian Gaussian Splatting [importance:35 dev:55]
- Algorithms for adaptive and heteroskedastic linear regression at the computational threshold [importance:15 dev:25]
- Random Hazard Forests [importance:15 dev:25]
- Very Exciting: Zero-Shot Model Predictive Control of Buildings via Excitation-Based Generalized Transfer Learning Models [importance:35 dev:45]
- Grouped Value Attention: Efficient KV Caching via On-Demand Key Reconstruction [importance:65 dev:75]
- Stochastic Gradient Descent over P2 [importance:25 dev:35]
- Real-World Deployment and Performance Characterisation of Fog-Based Deep Learning for Cold-Chain Temperature Prediction over LoRaWAN [importance:35 dev:55]
- Coaching Qwen3 Coder 30B to Think Like a CodeClash Arena Agent [importance:75 dev:85]
- Docker Containers vs. Virtual Machines: A Comparative Study of Architecture, Performance, Configuration, and Security [importance:45 dev:85]
- Models as Governed Interfaces for AI-Native MBSE: Read-Side Adequacy and Write-Side Admissibility [importance:55 dev:75]
- AgentGuard: Learning Execution Guardrails from Anomalous Coding-Agent Trajectories [importance:75 dev:85]
- Assurance Envelopes for Autonomous Coding Agents: Minimum-Cost Evidence for Software Change [importance:75 dev:85]
- Protocol-Preserving Context Trimming for Agentic Workflows: Benefits, Failure Regimes, and Budget Guardrails [importance:65 dev:85]
- AI Policies: Help or Hindrance? A Software Developer's Perspective [importance:45 dev:55]
- ExecuCritic: Calibrated Critic Shaping for Code Generation with Verifiable Rewards [importance:65 dev:75]
- An Exploratory Study of Dependabot Cooldown Adoption in Open-Source GitHub Projects [importance:55 dev:75]
- Memory-Skill Isomorphism: One Skill Carrier, Two Native Uses [importance:65 dev:85]
- RECTIFY: An Interactive Workbench for Post-Evaluation RAG Diagnosis, Repair, and Verification [importance:75 dev:85]
- RepoAtlas: Guiding Coding Agents via Evolving Multimodal Repository Views [importance:75 dev:85]
- TasmScan: Continuation-Aware Taint Analysis for TVM Bytecode with Savelist Abstraction [importance:55 dev:75]
- Search-Based Metamorphic Testing of Vision-Language Models in Autonomous Underwater Robotic Software [importance:45 dev:65]
- GANADI: Uncovering C/C++ OSS Reuse Genealogies via Pivotal Function-Based Clustering to Enhance Supply Chain Security [importance:55 dev:75]
- A Set-Theoretic Evaluation Framework for Assessing Asset Administration Shell Instances: Towards Comparability and Suitability [importance:35 dev:55]
- Towards an Asset Administration Shell Maturity Model [importance:35 dev:55]
- Grounding SWE-Agent Decisions in Architecture-0 Design: Navigating Unknown Unknowns through Physical Mapping [importance:75 dev:85]
- A Memorization Floor for LLM Refinement of Decompiled Code [importance:55 dev:75]
- An Exemplar of a Digital Twin in Mechanical Engineering: Understanding Model Hybridization [importance:35 dev:55]
- After the Party: Governing What a Viral Agent-Skill Ecosystem Left Behind [importance:75 dev:85]
- Coding Agents Have Converged: Why the SWE-bench Leaderboard Can No Longer Order Its Top Entries, and What to Measure Instead [importance:75 dev:85]
- Cognitive Admission Control: Risk-Conditioned Assurance for Consequential Actions in Agentic Distributed Systems [importance:75 dev:85]
- Evaluating the NIST Bugs Framework Against CWE as a Successor for Automated Vulnerability Classification [importance:55 dev:75]
- Model checking of hyperproperties for high-level relational models [importance:45 dev:75]
- Fine-grained Approaches for Confidence Calibration of LLMs in Automated Code Revision [importance:65 dev:85]
- From Procedural Skills to Strategy Genes: Towards Experience-Driven Test-Time Evolution [importance:75 dev:85]
- OpenGame: Open Agentic Coding for Games [importance:65 dev:85]
- On the Reliability of Code Comprehension Proxies [importance:45 dev:65]
- Can AI be Easy? Lessons Learned from the EZR.py Toolkit [importance:45 dev:75]
- The Biomimetic Architecture of Software 4.0 [importance:55 dev:85]
- Specifications for Humans, Agents, and Tooling [importance:75 dev:85]
- MANGO: Automated Multi-Agent Test Oracle Generation for Vision-Language-Action Models [importance:55 dev:75]
- PyMETA: Evaluating Student Code Diagnosis on and Beyond the First Execution Error [importance:45 dev:65]
- XREPOTEST: Benchmarking Multilingual Repository-Level Unit Test Generation for Large Language Models [importance:65 dev:85]
- A Governance Methodology Layer for AI-Assisted Software Development: Defect Taxonomy, Controlled Ablation, and a Test of Process-Over-Capability [importance:75 dev:85]
- An Empirical Analysis of CodeQL False Positives and Query Refinements for Java Vulnerabilities [importance:55 dev:85]
- Scalable Benchmarking Framework for Dynamic Quantum Circuits [importance:35 dev:65]
- SoK: Post-Quantum Cryptography Implementation in Software: Approaches, Challenges and the PQC-HOT Framework [importance:65 dev:85]
- Measuring Curriculum Alignment across Topical Coverage, Competency, and Cognitive Depth: A Longitudinal Framework Applied to CS2013 and CS2023 [importance:35 dev:55]
- Understanding the (In)Security of Vibe-Coded Applications [importance:75 dev:85]
- Shared Selective Persistent Memory for Agentic LLM Systems [importance:75 dev:85]
- The Verifier is the Curriculum: Precision Sets the Return on Search in Code Self-Distillation [importance:65 dev:85]
- Trusting-Trust Attack against an Entire Linux Distribution through Binary Manipulation [importance:75 dev:85]
- ADeptS-Bench: Measuring the Trustworthiness of Computer Use Agents Across Devices [importance:75 dev:85]
- Moirae: A Multimodal Agent Collaborative Framework for Dynamic Android Malware Detection [importance:65 dev:85]
- API Benchmark Scores Do Not Reliably Transfer to Chatbot Interfaces [importance:75 dev:85]
- A/I Shuts Down [importance:0 dev:0]
- Forgery of C2PA on a Pixel 10 [importance:55 dev:75]
- Ubuntu 26.10 completes transition to Rust-based coreutils [importance:55 dev:85]
- Reinventing issue tracking: Local-first and Git-native [importance:55 dev:85]
- The smallest possible Linux distribution [importance:35 dev:75]
- GEFS on OpenBSD: A very early preview [importance:35 dev:75]
- Some things Veloren does differently [importance:25 dev:65]
- Maintaining the love for coding in the time of AI [importance:35 dev:55]
- How to get a DOI for your blog posts [importance:15 dev:35]
- Replacing Pull Requests with Delta [importance:45 dev:85]
- JDK 27 has been released [importance:45 dev:85]
- When adding a fractional part to a number fixes your shader [importance:25 dev:75]
- Simple and Efficient Row-Level Security [importance:55 dev:85]
- The end of verygoodsoftwarenotvirus.ru [importance:25 dev:55]
- Why building a Rust LSP is hard [importance:45 dev:85]
- Unicode 18.0.0 [importance:35 dev:75]
- Coreutils - rejected feature requests [importance:25 dev:75]
- type declaration syntax [importance:35 dev:85]
- Agent State in the Tmux Status Line [importance:45 dev:85]
- Swift 6.4 Released [importance:45 dev:85]
- OSRS Wiki and RuneLite are increasingly under strain from low-effort AI development [importance:45 dev:65]
- A Minimal AGENTS.md and Cursor Rules Setup for Next.js App Router [importance:0 dev:85] (いいね相当スコア: 0)
- I have designed libraries professionally for 5 years [importance:0 dev:75] (いいね相当スコア: 0)
- Lovable App Builder: Complete Funding Timeline & Founders [importance:0 dev:45] (いいね相当スコア: 0)
- I replaced coding-agent orchestration with Git worktrees and one Python file [importance:0 dev:85] (いいね相当スコア: 0)
- Portfolio Optimization ML: Proven Risk-Return Edge [importance:0 dev:55] (いいね相当スコア: 0)
- Top AI Security Tools for Enterprises (2026) 💎 [importance:0 dev:75] (いいね相当スコア: 5)
- We Audited 50+ AI Agent Systems — Here's What CISOs Keep Missing [importance:0 dev:85] (いいね相当スコア: 1)
- How Crypto Recovery Specialists Build an Investigation Strategy [importance:0 dev:20] (いいね相当スコア: 0)
- I gave Claude Code $100 and 30 days to make a profit. Day 1, it built a product. Here's the pattern it used. [importance:0 dev:85] (いいね相当スコア: 0)
- Why Is Digital Currency in Demand, and Why Is It Considered the Currency of the Future? 🌐💰 [importance:0 dev:20] (いいね相当スコア: 0)
- Case Study: Architecting Neocloud Compute for the MCP & Agentic Workload Era [importance:0 dev:85] (いいね相当スコア: 4)
- FAQ: Five Myths About Agent-Installed Dependencies [importance:0 dev:85] (いいね相当スコア: 0)
- NobodyWho vs Cactus compared on engine design, model format, hardware, platforms, cloud, and licensing. [importance:0 dev:85] (いいね相当スコア: 0)
- Changes to LLM pricing: Baidu, Inceptron, Morph, StreamLake and Tencent [importance:0 dev:65] (いいね相当スコア: 0)
- Autoregressive vs Diffusion: A Different Way AI Could Generate Text [importance:0 dev:75] (いいね相当スコア: 10)
- Every Agent Session Is a Test Run [importance:0 dev:85] (いいね相当スコア: 0)
- Wire an AI Agent Into n8n [importance:0 dev:85] (いいね相当スコア: 0)
- Changes to LLM pricing: Baidu, Inceptron, Io Net, Morph, NextBit and StreamLake [importance:0 dev:65] (いいね相当スコア: 0)
- Your LLM provider list is five lists, and they already disagree [importance:0 dev:85] (いいね相当スコア: 0)
- tiktoken vs count_tokens: My Claude Budget Was 17% Off [importance:0 dev:85] (いいね相当スコア: 0)
- Changes to LLM pricing: AkashML, Alibaba, Baidu, Inceptron, Phala and StreamLake [importance:0 dev:65] (いいね相当スコア: 0)
- Evals: I Stopped Asking Whether the LLM “Looks Good” and Started Measuring [importance:0 dev:85] (いいね相当スコア: 0)
- The AI handoff problem: why switching models wipes your work (and what actually fixes it) [importance:0 dev:75] (いいね相当スコア: 0)
- Running an AI Agent Locally: ADK, Gemma 4, and Docker Model Runner [importance:0 dev:85] (いいね相当スコア: 1)
- Claude 4.6 was peak and it's downhill since then [importance:5 dev:10]
- I recreated the viral riso animation with Claude Code + Opus 5. Here's the full prompt, the process and the token count [importance:15 dev:25]
- A lot of talk about reduced weekly limits but a lack of data - so here's some actual numbers [importance:35 dev:25]
- I just tried astra. I ran out of usage for the week in 3 hours [importance:10 dev:5]
- The new usage limits make subscription and team plans genuinely useless for real work [importance:35 dev:30]
- Anthropic says Claude for Small Business has reached 900,000 installations since launching in May [importance:40 dev:5]
- I'm a fully blind business owner. I just sold my first vibe coded product for $1700. [importance:20 dev:30]
- What I Imagine Hearing While Claude Code Grinds on a Question [importance:2 dev:5]
- My Claude usage suddenly dropped… anybody having same exp? [importance:15 dev:10]
- I tested 3 more Claude Code plugins to cut costs. Here’s my verdict [importance:30 dev:40]
- Insane how fast limits get eaten [importance:10 dev:5]
- Even Fable is done with Opus' verbosity... [importance:5 dev:10]
- Thoughts on the NEW unified Claude? [importance:30 dev:25]
- I had 100% usage used up, and now I noticed they gave me back 30%! [importance:15 dev:5]
- Claude Opus 5 drew every frame of this animation using JavaScript. [importance:25 dev:45]
- New usage limits [importance:35 dev:10]
- Time to drop to 5x [importance:10 dev:5]
- Three reasons your Claude quota burns faster this week. Only one was announced. [importance:50 dev:65]
- Quota draining? Claude Code re-bills a finished sub-agent's whole context on a follow-up [importance:60 dev:75]
- Built a Claude Code plugin that lets Claude run a full AWS security scan and explain the findings conversationally !! [importance:40 dev:75]
- I used Claude to write a CapCut replacement and now people are actually ditching CapCut for it. [importance:35 dev:50]
- No comment. [importance:0 dev:0]
- LARA: small, composable behaviours for frozen LLMs [P] [importance:45 dev:75]
- GoBench: Evaluating LLMs on the game of Go [R] [importance:35 dev:60]
- I trained a 44M parameter quantized LLM from scratch on 45B tokens. It ships in 19.8 MB and runs at ~1,900 tok/s on CPU. [P] [importance:55 dev:80]
- NeurIPS 2026: handling of multiple venue locations seems bad [D] [importance:5 dev:0]
- TabPFN-3.5 is released as the next SOTA tabular foundation model [N] [importance:50 dev:75]
- NeurIPS Reference Check Response[D] [importance:0 dev:0]
- How much work in progress can a workshop submission be [R] [importance:0 dev:0]
- [D] How do you get preprocessed dataset of a paper [D] [importance:15 dev:50]
- Quoting Mustafa Suleyman [importance:25 dev:20]
- Underwriting Superintelligence: Backing Agents you can Sue — Rune Kvist, AIUC [importance:35 dev:35]
- [AINews] Jev: a “System One Model” that only decides/classifies/routes/scores — >100x faster, >200x cheaper than small frontier LLMs [importance:60 dev:80]
- Can Skills Learned in Games Transfer to Real-World Work? [importance:45 dev:65]
- [AINews] AEF-1 standard emerges for Third Party Evaluators, as Xai, OpenAI, and Anthropic all cosign [importance:65 dev:75]
- Project Feed [importance:15 dev:55]
- PhraseVault 3.0 [importance:10 dev:40]
- Thread [importance:10 dev:25]
- Appwrite 2.0 [importance:25 dev:80]
- Expand Board for macOS [importance:10 dev:25]
- flat.social [importance:15 dev:35]
- Fide Island [importance:10 dev:30]
- PeekPaste [importance:5 dev:25]
- Portfolio Frame [importance:5 dev:30]
- Mac Duo [importance:3 dev:10]
- DeepSeek AI engineer slams Anthropic, OpenAI over ‘pacing’ calls, invokes Hitler - South China Morning Post [importance:25 dev:20]
- DeepSeek Engineer Stuck In Self Destructive Loop – Says Building AI To Replace Him Is The Only Way To Stop OpenAI & Anthropic From Creating A “Cyberpunk 2077” Society - Wccftech [importance:25 dev:20]
- Employee of China's OpenAI and Anthropic rival, DeepSeek, writes a 'sentimental letter' on AI; says: Huma - The Times of India [importance:25 dev:20]
- "Like Hitler Getting Nuclear Bomb": China's DeepSeek Techie Warns Of Anthropic Dominating Advanced AI - ndtv.com [importance:25 dev:20]
- A DeepSeek engineer's viral post on talent, power, and humanity is a rare snapshot of work at China's frontier AI labs - Business Insider [importance:25 dev:20]
- DeepSeek-V4.1-Flash & Hermes Boosted Resident Evil 7 Performance By 50% On A 4-Year-Old Smartphone With Upscaling And Reducing Textures By 68% Without Quality Loss - Wccftech [importance:20 dev:15]
- DeepSeek engineer says Anthropic AGI would be "like Hitler getting nukes" - Cybernews [importance:25 dev:20]
- Why a DeepSeek Engineer Is Helping AI Replace Him - Geopolitechs [importance:25 dev:20]
- The Sequence Learning Loop - Issue 934: Understanding DeepSeek V4.1 Flash, DeepMind’s AlphaGenome Atlas and Muse - TheSequence | Jesus Rodriguez [importance:35 dev:65]
- U.S. Plans Sanctions on Chinese AI Distillation Practices - 조선일보 [importance:35 dev:40]
- PI a Minimal Agent Toolkit Offers an Easy Way to Run Local AI Agents - Geeky Gadgets [importance:40 dev:75]
- "I Hope the One Who Revolutionizes Myself Is Myself": DeepSeek Engineers Embrace Fears Amid AI's Accelerated Advancement - 36Kr [importance:25 dev:20]
- China's DeepSeek engineer compares Anthropic controlling most advanced AI to Hitler - Anadolu Ajansı [importance:25 dev:20]
- How to Setup DeepSeek as the Default Model on ChatGPT Codex - HackerNoon [importance:20 dev:65]
- DeepSeek V4 Flash vs GLM-5.3 vs Qwen3.8 Flash: 53x Gap [2026] - tech-insider.org [importance:50 dev:80]
- AI Agrees with Wrong Answers 97% After 25 User Repetitions, Study Finds - 조선일보 [importance:45 dev:70]
- DeepSeek-V4.1-Flash Outpaces GPT-5.6 Sol on Some Agentic, Coding Tests - thelec.net [importance:55 dev:80]
- DeepSeek Engineer's Viral Post Sparks Fierce Debate: Top Operator Development Talent Mulls Career Transition - 36Kr [importance:25 dev:20]
- AI is not a monopoly China’s top newspaper says - asahi.com [importance:30 dev:25]
- DeepSeek Developer Compares Anthropic's AI Domination to Hitler Getting the Atomic Bomb First - news.sbs.co.kr [importance:25 dev:20]
- DeepSeek engineer compares Anthropic's control of AI to Hitler obtaining atomic technology before the Allies - UA.NEWS [importance:25 dev:20]
- DeepSeek engineer likens Anthropic AI dominance to Hitler’s nuclear threat - news.az [importance:25 dev:20]
- DeepSeek Open-Sources Harness Agent Runtime With Everything-Is-a-Plugin Design - Pandaily [importance:50 dev:80]
- China's AI Adoption Rate Surpasses 50%: Key Statistics and Industry Insights - 36Kr [importance:30 dev:20]
- Liang Wenfeng Appointed as CFO: Former Investor in Zhipu AI and MiniMax Takes New Position - 36Kr [importance:20 dev:10]
- China Rebuts 'AI Slowdown' Debate: 'Ruthlessly Crush' and 'Atomic Bomb' Warnings Revealed - news.sbs.co.kr [importance:25 dev:20]
- After Launching V4.1 Core Operator, Is DeepSeek’s Senior Operator Engineer Shifting Gears? Late-Night Monologue on Tech Development & Career Insights - 36Kr [importance:25 dev:20]
- DeepSeek Engineer Sounds AI Alarm: Not Worried About Losing His Job, But He Must Switch Tracks - finance.biggo.com [importance:25 dev:20]
- Robi Axiata launches 'AI Space' to provide direct access to global AI models via MyRobi App - telecompaper.com [importance:25 dev:30]
- Shady Capital Mastermind Behind Company A's "AI Doomsday" Hype Exposed: Stoking Public Anxiety for IPO Fundraising & Hostility Toward DeepSeek's Open-Source Initiatives - 36Kr [importance:20 dev:15]
- Will AI kill humans? ThePrint asked five AI chatbots & their answer is neither ‘yes’ nor ‘no’ - ThePrint [importance:20 dev:25]
- DeepSeek Appoints New CFO: Key Leadership Update Marks a Major Milestone - 36Kr [importance:20 dev:10]
- Anthropic with best AI is like Hitler with nuclear bomb, DeepSeek engineer says communism only viable future - India Today [importance:25 dev:20]
- China asks to "prepare before it rains" in the face of AI risks - Diari ARA [importance:30 dev:25]
- 9 Things to Know About DeepSeek’s New 552B AI Model - Analytics Insight [importance:35 dev:40]
- DeepSeek taps first CFO ahead of IPO to step up low-cost AI push - 디지털투데이 [importance:20 dev:10]
- Grok Imagine Image 2.0 API: 12 Steps [2026] - tech-insider.org [importance:20 dev:60]
- xAI Continuing To Try To Block Minnesota’s New AI “Nudification” Law - knoxradio.com [importance:25 dev:20]
- SEOAgent Releases a New Software Update for Coding Agents, Adding Grok Bot Support and Open Knowledge Format Publishing to Improve Visibility in AI Search - TMX Newsfile [importance:30 dev:65]
- Ongoing Study Reveals Epistemological Flaws in Gemini and Grok as Risk Factors for AI Safety and Alignment - PR Newswire [importance:40 dev:70]
- The world's richest man says he's living in a trailer - Business Insider [importance:10 dev:5]
- Elon Musk’s xAI resolves claims against Apple over AI competition - mercurynews.com [importance:20 dev:10]
- How to Stop Grok From Training on Your X Posts and Chats - Techloy [importance:20 dev:30]
- When treated as therapy clients, AI chatbots generate elaborate narratives of trauma and punishment - PsyPost [importance:30 dev:45]
- Mark Cuban Asked Grok Whether AI Or Climate Change Will End Humanity. It Didn't Hesitate - Benzinga [importance:20 dev:20]
- Elon Musk’s xAI Asks Appeals Court to Block Minnesota Law Targeting AI-Generated Nude Images | MinneapoliMedia - minneapolimedia.town.news [importance:25 dev:20]
- Musk urges top AI labs, Chinese companies to test each other's models amid calls for slowdown - cnbc.com [importance:30 dev:25]
- Grok AI Predicts that Ethereum Could Hit $12,000 by the End of 2026 - 99Bitcoins [importance:15 dev:15]
- xAI Data Center Noise Controversy, Explained - BASENOR - Tesla Accessories [importance:20 dev:10]
- Elon Musk Admits AI Isn’t Good Enough For “Extremely High-Performance Software” & Says Grok 4.9 Should Match Fable Class Models - Wccftech [importance:25 dev:30]
- Elon Musk's New Airstream Home Near xAI Project Expansion - GuruFocus [importance:10 dev:5]
- xAI Lawsuit Could Have Massive First Amendment Implications - trillmag.com [importance:25 dev:25]
- Philippines Must Speak With One Voice On China – OpEd - Eurasia Review [importance:10 dev:5]
- Artificial Intelligence And Biosecurity Issues – Analysis - Eurasia Review [importance:35 dev:40]
- Musk Reveals Grok 4.8’s Pre-Training Stack is Written in C++ by ‘Humans’, Not AI - analyticsindiamag.com [importance:20 dev:50]
- OpenCode launches Union Alpha model for free use on OpenRouter - Crypto Briefing [importance:50 dev:70]
- AI修正パッチの「成功率26%」をどう読むか：Agent評価の5つの罠 [importance:0 dev:80] (いいね相当スコア: 1)
- 比較】BeautifulSoup自作 vs Scraping AI：開発者が知るべき「保守税」と使い分けの境界線 [importance:0 dev:85] (いいね相当スコア: 0)
- 🔒クラウドに一切送らないローカルAIアシスタントを、個人開発の副産物として作った話 [importance:0 dev:80] (いいね相当スコア: 1)
- LLMは「正答率」だけ見ればいい？ ―― RAG・Agent・Fine-tuningを評価する方法 [importance:0 dev:80] (いいね相当スコア: 0)
- LLMにWikiを書かせて半年、一番役に立った画面はLLMの文章を使っていなかった [importance:0 dev:75] (いいね相当スコア: 6)
- 無料のAIを自分のVPSで動かす。Ollamaで作る「自分専用AI環境」入門 [importance:0 dev:80] (いいね相当スコア: 0)
- LLM APIの履歴管理とContext Caching [importance:0 dev:85] (いいね相当スコア: 0)
- 【PyTorch実装】ルーローの三角形に学ぶ：MoEルーターの「適応型偏心ルーティング」による動的スパース化と計算量削減 [importance:0 dev:80] (いいね相当スコア: 0)
- Kimi K3 と DeepSeek V4.1 Flash を実務タスクで比較 — 正しさは互角、使い所は別だった [importance:0 dev:75] (いいね相当スコア: 0)
- YANS2026 参加レポート [importance:0 dev:60] (いいね相当スコア: 0)
- LLMによる求人のコールドスタート推薦の留意点 [importance:0 dev:65] (いいね相当スコア: 2)
- LLMを業務に入れて分かった「オントロジー」の効き方 — 運に頼らないAI運用の骨組み [importance:0 dev:85] (いいね相当スコア: 3)
- Claude Code 101を日本語で#1 Claude Codeとは何か｜エージェントループの仕組みと最初の一歩 [importance:0 dev:90] (いいね相当スコア: 0)
- AIで採用スクリーニングする前に — 個人情報を守る匿名化前処理パイプライン [importance:0 dev:85] (いいね相当スコア: 1)
- 学習データを汚染する？OWASP LLM04 Data and Model Poisoningを初心者向けに解説 [importance:0 dev:85] (いいね相当スコア: 1)
- AIエージェントの回答ばらつきを減らす：BigQuery Graphによる候補制約 [importance:0 dev:85] (いいね相当スコア: 1)
- AIに考えることを任せすぎていないか [importance:0 dev:45] (いいね相当スコア: 0)
- Tech Watch 2026-09-15: 評価と運用が主役になる [importance:0 dev:65] (いいね相当スコア: 0)
- Gemma 4 31Bの重複計算を解消し、2枚のGPUでdecode 30 tok/sへ [importance:0 dev:85] (いいね相当スコア: 0)
- タスクの優先順位付けLLMは、情報が足りなくても質問せず推測で埋めて自信満々に答える [importance:0 dev:80] (いいね相当スコア: 0)
- 数字が上がっていく ― 定点観測と地道な調整、そしてその数字を疑う ― KotobaCore開発秘話（5/8） [importance:0 dev:85] (いいね相当スコア: 0)
- Jevのconfidenceは何を測っているのか ── 公式サンプル6件から算出式を逆算した [importance:0 dev:85] (いいね相当スコア: 1)
- DPOを4本のlog probabilityから実装する [importance:0 dev:85] (いいね相当スコア: 0)
- Self-paced Ensemble(ICDE 2020)による不均衡データの分類 [importance:0 dev:80] (いいね相当スコア: 0)
- 村中仁斗（むらなかまさと）｜G検定合格体験記｜AI・機械学習を学んで資格取得 [importance:0 dev:35] (いいね相当スコア: 0)
- 【音声認識】リアルタイム話者分離の評価について [importance:0 dev:75] (いいね相当スコア: 0)
- リークは3回、姿を変えて現れた、、陽性 n=2 の振動異常検知で踏んだ、分割・採点式・モデル選択のリーク [importance:0 dev:85] (いいね相当スコア: 1)
- 初めて事前学習モデルを作った！車両の輪郭を塗る「JALO」の開発記録 [importance:0 dev:85] (いいね相当スコア: 0)
- AIが東大に受かる時代に、AI五輪メダリスト高校生は何を作ったか [importance:0 dev:65] (いいね相当スコア: 2)
- プライマー設計・温度条件・比色判定——LAMP検査パイプラインの入口から出口まで、AIが入り込んできた話 [importance:0 dev:55] (いいね相当スコア: 0)
- Autoware Robotaxiの現在地（2026-09） [importance:0 dev:75] (いいね相当スコア: 1)
- 個人実証(まとめ)：PoCを実運用にどう接続するか（技術的証明と組織的意思決定の間） [importance:0 dev:85] (いいね相当スコア: 1)
- 予測精度54.8%のAIは利益を出せるのか？WFOから売買戦略まで検証した | 第8回：AIで為替の未来予測は本当にできるのか？ [importance:0 dev:75] (いいね相当スコア: 0)
- 発走5分前、オッズの旅はまだ6割しか終わっていない — 13日間の時系列観測で見えた「情報の到着時刻」 [importance:0 dev:75] (いいね相当スコア: 0)
- RLHFを選好データ・報酬モデル・PPOから実装目線で整理する [importance:0 dev:85] (いいね相当スコア: 0)
- 固定学習率のまま確率的勾配降下法が収束する条件 [importance:0 dev:85] (いいね相当スコア: 0)
- Google Teachable Machineから入門するニューラルネットワーク [importance:0 dev:75] (いいね相当スコア: 0)
- 「陽性だから病気の確率が高い」はなぜ間違うのか [importance:0 dev:60] (いいね相当スコア: 0)
- BigQueryサービスのAI機能・関数まとめ①[整理編](2026/07) [importance:0 dev:85] (いいね相当スコア: 5)
- macOS 27にはローカルLLMが入っている [importance:0 dev:85] (いいね相当スコア: 0)
- LLMの「覚えている」を信用しない — ECHO AgentのIdentity Continuity Gate設計 [importance:0 dev:85] (いいね相当スコア: 0)
- NotionのAgent SkillsをClaude Code/Cursorに共有する新機能まとめ [importance:0 dev:80] (いいね相当スコア: 0)
- Jevってなんだ？ — 文章を書かないAIに、LLMのif文判定を任せられるか [importance:0 dev:85] (いいね相当スコア: 0)
- 連続時間マルコフ過程とDiscreate Diffusionメモ [importance:0 dev:85] (いいね相当スコア: 0)
- 村中仁斗（むらなかまさと）｜G検定に合格した勉強法・学習記録｜AI・機械学習 [importance:0 dev:35] (いいね相当スコア: 0)
- 【技術解説】【完全ガイド】PythonでEDINET APIを使った金融データ取得とエラー解決 [importance:0 dev:75] (いいね相当スコア: 0)
- 「○○を作って」と言うだけ。ローカルLLMがClaude Code / Codexを自動で使い分ける環境を作る【初学者向け・Windows対応】 [importance:0 dev:90] (いいね相当スコア: 取得失敗)
- AIチャットに読ませた著作物は学習されるのか｜「読む」と「覚える」はまったく違う [importance:0 dev:45] (いいね相当スコア: 取得失敗)
- 【第4章：LLMの正体編】第31話：「最後は確率で決めるの！？」LLMが“次のToken”を選ぶまで [importance:0 dev:75] (いいね相当スコア: 取得失敗)
- 自由文の生成を捨てた「System One Model」とJevは何を変えるのか [importance:0 dev:85] (いいね相当スコア: 取得失敗)
- 一人二役の台本を、一人で読み上げるAI [importance:0 dev:65] (いいね相当スコア: 取得失敗)
- 記憶喪失になったGeminiとその原因を議論してみた [importance:0 dev:55] (いいね相当スコア: 取得失敗)
- AIパートナーは人間関係を減らすのか？――約1年の追跡研究 [importance:0 dev:35] (いいね相当スコア: 取得失敗)
- 🔊音声あり（日＆英）：【AIセキュリティ最新論文】LLMのハリーシネーションを防げ！「モデルの提案とコードの決裁」による脆弱性診断の裏側 [importance:0 dev:85] (いいね相当スコア: 取得失敗)
- 【生成AIニュース+】『Jev』『Meshy 7.1』『LynnReal-Omni』『Flet 1.0』『AlayaVista』『MuLaCover』『LM Studio Bionic 1.1.3』『BrowserSkill』『Xiaomi-CocktailASR-1』『G-ray』『MiniMax-H3-Longvideos』『Over the Reality』『PixiJS 3D』『ComfyUI-FL-YuE2』『ComfyUI H3 Motion Context MultiRef』他 [importance:0 dev:55] (いいね相当スコア: 取得失敗)
- #9 全部をLLMに投げるのを、やめる｜System Oneモデル「Jev」が返す3つの型 [importance:0 dev:85] (いいね相当スコア: 取得失敗)
- AIの知識はいつの時点で止まっているのか？――学習済み知識・検索・RAGの違い [importance:0 dev:60] (いいね相当スコア: 取得失敗)
- 2026-09-16 Hacker News Top 10 [importance:0 dev:65] (いいね相当スコア: 取得失敗)
- 続・学歴の価値を考える [importance:0 dev:0] (いいね相当スコア: 取得失敗)
- LLMは膨大な百科事典にすぎない [importance:0 dev:75] (いいね相当スコア: 取得失敗)
- 【論文】【AI】限られたデータを何度も学ぶ [importance:0 dev:85] (いいね相当スコア: 取得失敗)
- ハルシネーションを考える時にやりがちな事 [importance:0 dev:75] (いいね相当スコア: 取得失敗)
- ChatGPTの元開発者が仕掛ける！「Jev」が目指す意思決定AIの新世界 [importance:0 dev:85] (いいね相当スコア: 取得失敗)
- 生成に安全を“織り込む”検査の新段階 [importance:0 dev:85] (いいね相当スコア: 取得失敗)
- Codexの使用量を減らすために、ローカルLLMを「司令塔」にしたらどうなる？ [importance:0 dev:85] (いいね相当スコア: 取得失敗)
- 同じモデルで42%と78%。エージェントの性能を決めるハーネスの設計と、自作とマネージドの分岐点 [importance:0 dev:90] (いいね相当スコア: 取得失敗)
- AIと共に生きる。日々の開発とクリエイティブを支えるローカルAI環境の裏側 [importance:0 dev:80] (いいね相当スコア: 取得失敗)
- エッジAI・オンデバイスの現在地：ローカルで動くLLMが開発や日常をどう変えるか [importance:0 dev:75] (いいね相当スコア: 取得失敗)
- Googleが「Gemini 3.8 Live」をリリース！リアルタイム対話型コーディングの実力を徹底解剖 [importance:0 dev:85] (いいね相当スコア: 取得失敗)
- AIが数学の未解決問題を解いた、と聞いて浮かれた翌週｜「まだ夢見るには早い」と冷静に指摘した記事の話 [importance:0 dev:80] (いいね相当スコア: 取得失敗)
- Devinが仮想環境でmacOSの提供開始。Macの実機不要でDevinがコード生成、テスト、デバッグ、実行、AppStore配信前のベータ公開まで実行 [importance:75 dev:90]
- Android、パスキーをパスワードマネージャ間で転送可能に。Googleパスワードマネージャ、1Password、Bitwarden、Dashlaneなどが対応 [importance:20 dev:45]
- 「Java 27」正式リリース。全環境でG1 GCがデフォルトに、TLS 1.3用に耐量子暗号のハイブリッドキー交換など新機能 [importance:20 dev:55]
- オラクル、Javaのセキュリティパッチを毎月提供開始、AIが脆弱性の発見を加速しているとして [importance:20 dev:45]
- 個人のAIエージェントを、チームの力に変えるまで 〜Yahoo!検索のAX実践〜 [importance:70 dev:85]