AI News

REPAIR: Resolving Long-Tail Confusion in Scientific Retrievers via Fact-Verified Iterative Refinement (arxiv.org)

Re2A: Situated Conversational Recommendation via Rubric-based Preference Reasoning and Alignment (arxiv.org)

Characterizing Replay Retention Under Dynamics Shift in Model-Based Reinforcement Learning (arxiv.org)

SAiFE-gym: Model-based Environments for Automated Market Making with Concentrated Liquidity (arxiv.org)

Derivative-Free Structured Updates for Muon (arxiv.org)

Similarity Pairing with Energy Mover's Distance for Self-Supervised Pre-Training at the LHC (arxiv.org)

Machine learning kinetics from molecular dynamics data (arxiv.org)

FAME: An FPGA-Based Platform for Approximate Multipliers Evaluation with Pattern-Guided DNN Retraining (arxiv.org)

WFM: Wiki Foundation Model for Complex Agentic Reasoning (arxiv.org)

Time-Aligned Evolving Concept Graphs for Scientific Relation Forecasting (arxiv.org)

Symbolic Temporal Supervision of LLM Agents Using Contracts (arxiv.org)

Designing Agentic AI Workflow Portfolios under Imperfect Selection and Compute Cost (arxiv.org)

AutoTuneBench: Trustworthy Measurement for Agent Auto-Tuning of LLM Serving Engines (arxiv.org)

Beyond Direct Sensing: Harnessing Indirect Observations from Third-Party Sensors in Vehicle Tracking (arxiv.org)

Self-Supervised Learning for Robust Resonance Mass Regression in Cascade Decays (arxiv.org)

Decoding Extrahepatic Targeting of Lipid Nanoparticles with Interpretable Machine Learning (arxiv.org)

Robust and Efficient AI Frameworks for Scalable Material Design and Property Prediction (arxiv.org)

Rethinking Domain Specialization for Open-Ended Scientific Reasoning in Astronomy Language Models (arxiv.org)

Democratizing Clinical Tumor Whole Genome Sequencing: 18-hour End-to-end Analysis via Trillion-parameter Large Language Models Locally Deployed on Consumer-grade Hardware (arxiv.org)

When Is Graph Structure Worth Its Cost? The Case for Structure Pricing in Retrieval-Augmented Generation (arxiv.org)

Decodability is Not Causality: Dissociating Probe Readouts from Behavioral Drivers via SAE Decomposition (arxiv.org)

Teaching AI, Robotics, & Community: A Hubs-Based K-12 Education Framework for Reaching Rural Schools (arxiv.org)

The Other Half of the Memory Wall: Serving 35B MoEs from SSD with Trained Routing Prediction (arxiv.org)

Anchoring What Matters: A Dual-Level Learning Framework for Visually-Grounded Multimodal Reasoning (arxiv.org)

Behavior2Value: Benchmarking and Empowering LLMs for Consumer Value Measurement from E-commerce Behaviors (arxiv.org)

Missing Bridges: Composition-Aware Active Imitation Learning (arxiv.org)

Multimodal Conditioning of Fine-Tuned Stable Diffusion XL for Controllable and Culturally Faithful Ulos Motif Generation (arxiv.org)

RideWay: Benchmarking Efficient Task Completion for Tool-Using Language Agents (arxiv.org)

TuiML: Machine Learning for AI Agents (arxiv.org)

Reinforcement Learning for Real-Time Vision-Language-Action Policies (arxiv.org)

Timbre Analysis of the Hulusi, a Southwestern Chinese Free-Reed Instrument, using Machine Learning (arxiv.org)

Making Political Text Scaling Comparable: Infrastructure and Hyperparameter Sensitivity for 17 Algorithms (arxiv.org)

Generic Characteristic-Zero Equivalence Between Derivative B\'ezout Inversion and Multipoint Evaluation (arxiv.org)

GroupKV: Hierarchical KV Cache Management for Long-Context Diffusion LLM Inference (arxiv.org)

Contiguity, Not Importance: Budgeted Repair of Stale KV Caches After Document Edits (arxiv.org)

When to Call an LLM: A Confidence-Gated Hybrid for Cost-Effective Emotion Recognition in Conversational AI (arxiv.org)

Memory Has Geometry: Non-Uniform Geometric Memory for Long-Horizon Personalized AI (arxiv.org)

Measuring AI Leadership: Development and Validation of a Multidimensional Measure for AI-Native Organizations (arxiv.org)

Collaborative Memory for Multi-Agent VLM Systems (arxiv.org)

OBC-Prune: Outcome-Based Calibration for Large Reasoning Model Pruning (arxiv.org)

ERPBench: A State-Grounded Evaluation Paradigm for Computer-Use Agents in Enterprise Software (arxiv.org)

Acting in Meters: Learning Metric Interactions for Precise Robotic Manipulation (arxiv.org)

BLADE: ReliaBle Dynamic Hardware-Aware SNN-ANN Boundary SeLection for Event-BAseD Object DEtection (arxiv.org)

WARD: Runtime Workload-Adaptive Vision TRansformer Framework for Dependable Edge AI (arxiv.org)

Attention Dispersion as a Diagnostic Signal for Hallucination in Large Language Models (arxiv.org)

REQAP: Resilient Weight Packing and Quantization for Edge DNN Acceleration (arxiv.org)

No Usable Linear "Capitulation Direction" in Two Small LLMs: A Validation Protocol for Activation-Steering Claims, and a Cross-Family Behavioral Study of Sycophancy Under Pushback (arxiv.org)

Enhancing Extubation Failure Prediction with LLM-Derived Features from Respiratory Therapy Clinical Notes (arxiv.org)

MLLMs Hallucinate when Information Distribution Drifts in Synergy Heads (arxiv.org)

RecMorph: Topology-Guided Spatial Recurrence for Generalized Morphology Control (arxiv.org)