AI News

Accuracy Is Not Enough: A Cross-Architecture Audit of Demographic Bias in Deep Knowledge Tracing (arxiv.org)

How Far Can Sub-3B Open Language Models Go in Zero-Shot Essay Scoring on an 8 GB Consumer GPU? (arxiv.org)

Improving Online Reinforcement Learning via Bidirectional Behavior Prior Distillation (arxiv.org)

Placement Is Free, Composition Is Not: The Latin Square as a Provably-Balanced Construction for Heterogeneous Sequence-Mixer Stacks (arxiv.org)

Labeled Incidence Structures for Native Transformer Modeling of Text, Knowledge Graphs, and Hypergraphs (arxiv.org)

Sharp Reconstruction Bounds for Autoencoders Using the Same Forward Map (arxiv.org)

E-AVI: Evidence-Grounded Multimodal Assessment for Automated Video Interviews (arxiv.org)

Geopolitical Divisions Across Languages in Large Language Models (arxiv.org)

FedeRICo: Federated Region-Influenced Coupling for Traffic Flow Prediction (arxiv.org)

Can Data Attribution Filter Out Subliminal Learning? Not Reliably (arxiv.org)

DART: Distillation-Aware Reparameterization for Training-Free LoRA Reuse in Few-Step Video Diffusion Models (arxiv.org)

Personalising a Cross-User Surface Electromyography Encoder Under a Small Calibration Budget (arxiv.org)

Intact-to-Amputee Transfer in Surface-EMG Gesture Decoding: Training Source and Calibration Budget (arxiv.org)

A Learning Algorithm for Threshold Boolean Networks with Prescribed Fixed Points (arxiv.org)

Improving Generalization and Robustness in Offline Reinforcement Learning via Boundary-Aware Data Augmentation (arxiv.org)

MAGMA-GEN: Validated Recovery Supervision from Ambiguous Failures via Counterfactual Re-Execution (arxiv.org)

WiCleanData: Guaranteeing the Type Consistency of Wikidata by Taxonomy Refinement and Constraint Enforcement (arxiv.org)

FCA-Guided Counterfactual Explanations for Multi-Modal Breast Cancer Diagnosis: A Framework Achieving Perfect Validity with Emergent Sparsity (arxiv.org)

Hypernetwork-Parameterized Spatially Adaptive Neural Operators for PDE Learning (arxiv.org)

ZeroHAT: Behavior-Conditioned Zero-Shot Human Activity Trace Generation (arxiv.org)

SAGG: Sample-Adaptive Gradient Gating for Robust Multimodal Learning under Heterogeneous Corruption (arxiv.org)

Marginal utility, matrix factorization, and the Key-Value (KV) cache: a unified information-economic framework for sovereign geo-mining inference (arxiv.org)

Tailored to you: longitudinal effects of personalising language models (arxiv.org)

A Proposal for an Agentic AI Architecture to Support Multi-Domain Decision-Making in the Brazilian Armed Forces (arxiv.org)

UnifiedPlayers: Enhance Tool-Integrated Reasoning in Agentic Reinforcement Learning (arxiv.org)

EviRec: Continual Evidence Learning for Dual Cold-Start POI Recommendation (arxiv.org)

Minimax-Optimal Online Contract Design with Unrestricted Bounded Contracts (arxiv.org)

Learning Principal-Agent Contracts for Equitable Smallholder Carbon Farming under Moral Hazard and Adverse Selection (arxiv.org)

SCGFM-ART: Amortized Relational Transport for Structure-Centric Graph Foundation Models (arxiv.org)

Solving Minimum Span Antibandwidth and Cyclic Antibandwidth Labeling Problems (arxiv.org)

Perception, Layout, and Validation: Calibrated Confidence for Reliable Straight-Through Processing of Financial Documents (arxiv.org)

MTVA-Bench: Evaluating the Language Model Inside Cascaded Voice Agents (arxiv.org)

PaGNet: A Panel-Aware GBDT--Neural Network for Multi-Target Corporate Tax Avoidance Proxy Forecasting (arxiv.org)

Sequential Contextual Fit Predicts Human Behavioural and Neural Dynamics Across Domains (arxiv.org)

Seismic Site Response Prediction from Sparse Observations Using Finite-Element-Pretrained Latent Dynamics (arxiv.org)

Training Neural Networks to Approach the Optimum Bayes Estimator in Dense Multi-Emitter Localization (arxiv.org)

Deep Learning-Based Classification of Cognitive and Resting States Using Electroencephalography Signals (arxiv.org)

Distributionally Robust Federated Learning with Multi-Source Data (arxiv.org)

JointMatch: A Unified Heterogeneous Graph Neural Solver for Large-Scale Ride-Sharing Matching (arxiv.org)

When AI Agents Commit: Cognitive Serializability Across Data, Evidence, Policy, and Authority (arxiv.org)

What Do We Expect from LLMs? Mapping the Design of LLM Benchmarks (arxiv.org)

Position: It is Time to Virtualize Foundation Models with a Self-evolving Operating System Layer (arxiv.org)

What Do Current Systematic Generalization Tasks Miss? A Reasoning-Centered Analysis (arxiv.org)

Radio Frequency Detection and Classification of Microplastics in Water (arxiv.org)

When EOS Tokens Disagree: Understanding Length Inflation in On-Policy Distillation (arxiv.org)

Mitigating Retaliatory Algorithmic Collusion in Repeated Games (arxiv.org)

CrystalMO-TuRBO: Multi-Objective Trust-Region Bayesian Optimization for High-precision Joint Crystal Structure Refinement (arxiv.org)

Characterizing Web Search by Conversational LLM Agents: From Search Decisions and Strategies to Results and Responses (arxiv.org)

Do AI Agents Understand Computer Architecture? (arxiv.org)

MAGS: Multi-agent Auto-formalization Guarantees Safety for Agentic Outputs (arxiv.org)