AI News

Towards Optimizing SQL Generation via LLM Routing (arxiv.org)

Beyond Quacking: Deep Integration of Language Models and RAG into DuckDB (arxiv.org)

Generalization Can Emerge in Tabular Foundation Models From a Single Table (arxiv.org)

Physically Aware Radiomics Without Interpolation: Disentangling Voxel Geometry and Signal Modification in CT and MRI (arxiv.org)

Natural-Language to SysMLv2 Translation via Conformance-Driven Iterative Refinement (arxiv.org)

Multilingual Agent System for Inclusive Wildfire Evacuation Guidance (arxiv.org)

From Semantic to Token Communication: The Next Paradigm for Large-Model-Driven 6G Intelligent Connectivity (arxiv.org)

Is Bash All You Need? An Empirical Study of Tool Interfaces for Enterprise Digital Worker Agents (arxiv.org)

Planning as Dynamics Relaxation: Hippocampal Recurrent Network Realizes Optimal Goal-Directed Navigation (arxiv.org)

ShieldVLA: Feasibility-Aware Safety Alignment for Vision-Language-Action Models (arxiv.org)

Evaluation of MLLM-Agnostic Plug-and-Play Keyframe Selection Methods for Long Video Understanding (arxiv.org)

An Evolutionary Computation Framework for Multi-Agent Q-Learning with Mean-Field Environmental Feedback (arxiv.org)

(How) Do MLLMs Report Bistable Images Like Humans? (arxiv.org)

Sampling headroom is not selection gain: a compute-value audit of test-time scaling for video world models (arxiv.org)

ReCAST: Reward Credit Assignment across Timesteps for Online Diffusion Reinforcement (arxiv.org)

TryOnReward: Learning Foveated Consistency for Reinforcement Fine-Tuning of Virtual Try-On (arxiv.org)

From Process Loss to Assembly Bonus: Human-Grounded Diagnosis of Multi-Agent LLM Collaboration (arxiv.org)

BEACON: Behavior and Appearance Control for Subject-Specific Video Generation (arxiv.org)

Forward-Facing Near-Infrared Adds Little to Colour for Farm-Machinery Traversability: A Site-Disjoint Evaluation of Sensor-Dependent Spatial Leakage (arxiv.org)

Conflict-Predictive Variable Horizons in Multi-Drone Distributed Model Predictive Control (arxiv.org)

Multimodal-Multiresolution Foundation Model for Lunar Remote Sensing (arxiv.org)

LLaDA-UI: Bringing Block-wise Diffusion to Vision-Language GUI Agents (arxiv.org)

Don't Count the Edits, Judge by the Outcome Alone: Reward-Based Evaluation for Grammatical Error Correction (arxiv.org)

Specification Oracles (arxiv.org)

When Greedy Sampling Explores: KL-Regularized Contextual Bandits without Eluder-Dimension Dependence (arxiv.org)

A Multi-Resolution Multi-Domain Pre-Training Framework for Universal Traffic Forecasting (arxiv.org)

Lie to me: Detecting Managerial Evasiveness in Earnings Calls via Conversational Audio Encoders (arxiv.org)

Machine Learning under Imperfect Data: Challenges and Methods (arxiv.org)

Learning Through Energy Refinement and Manifold Projection: A Cooperative EBM-AE Framework (arxiv.org)

Machine Learning in Fish Farming (arxiv.org)

Finite-Time Node Separation in Recurrent Graph Neural Networks with Persistent Gaussian Perturbations (arxiv.org)

AgentKV: Phase-Aware KV Eviction for Agentic LLMs (arxiv.org)

Minibatch persistency, eight years later: what batch reuse costs in steps and joules, and what it saves in data (arxiv.org)

Exploring napping paradigm for Recurrent Spiking Neural Networks (arxiv.org)

Optimal Transport for Efficient, Unsupervised Anomaly Detection on Industrial Data (arxiv.org)

Efficient Online Inverse Optimization with $O(d)$ Regret (arxiv.org)

Learning to Solve Hard Problems in RL for LLMs by Never Giving Up (arxiv.org)

A Machine Learning API for Earth Observation Data Cubes Based on openEO (arxiv.org)

On the Potential of Multi-Task Learning in Predictive Process Monitoring (arxiv.org)

One Spectrum, Two Resources: Data-Memory Scaling in Autoregressive Prediction (arxiv.org)

Pretraining for Sample-Efficient Neural Interfaces (arxiv.org)

A Three-Axis Stress Test of LLM vs Classical ML for Network Intrusion Detection under Distribution Shift and Adversarial Evasion (arxiv.org)

Adaptive Phase-Switching for Communication-Efficient Federated LoRA Fine-Tuning (arxiv.org)

Know When to Stop, Where to Restart: Accelerating Multi-Turn Agentic On-Policy Distillation (arxiv.org)

Mind2Dialogue: Training Human-Aware Language Models by Simulating User Mental States (arxiv.org)

Disentangling Representation Evolution in Transformers through Directional Decomposition (arxiv.org)

Mitigating the Stability-Plasticity Dilemma in Adaptive Train Scheduling with Curriculum-Driven Continual DQN Expansion (arxiv.org)

Hardware-Aware Learned Representation Compression for Distributed In-Sensor Vision (arxiv.org)

Tabby: An Open Pretraining Recipe for Time Series Foundation Models (arxiv.org)

Data-Efficient Agentic Graph Domain Adaptation via Reliability-Aware Prototype Learning (arxiv.org)