AI News

Score Function Gradient Estimation to Widen the Applicability of Decision-Focused Learning (arxiv.org)

MACCA: Offline Multi-agent Reinforcement Learning with Causal Credit Assignment (arxiv.org)

Multimodal Approaches for Visually-Rich Document Type Classification: A Comparative Analysis (arxiv.org)

Order within Chaos: Capturing Intrinsic Energy Anomalies for AI-Manipulated Image Forgery Localization (arxiv.org)

Consistency Training while Mitigating Obfuscation via Rate Matching (arxiv.org)

Nonlinear Equilibrium Transitions in a Potential Game Model for Federated Learning (arxiv.org)

VERA: Variational Inference Framework for Jailbreaking Large Language Models (arxiv.org)

Canonicalized Stable-List Replay for Private Federated Continual Learning over Language-Model Embeddings (arxiv.org)

Finer Parameter Steps for Low-Rank PEFT: A Controlled Study with CP Tensor Adapters (arxiv.org)

Auditing Near-Optimal Policies Can Be Exponentially Hard: Conditional Query Lower Bounds via Occupancy Rashomon Capacity (arxiv.org)

Dynamic Proxy-Mixing: Transferring Replay Controllers from Small to Large Models for Continual Instruction Tuning (arxiv.org)

Perturbation Effects on Accuracy and Fairness among Similar Individuals (arxiv.org)

Random Erasing vs. Model Inversion: A Promising Defense or a False Hope? (arxiv.org)

Towards Stable, Globally Expressive Graph Representations with Laplacian Eigenvectors (arxiv.org)

Implicit Regularization for Multi-label Feature Selection (arxiv.org)

A Foundation Model for Wearable Movement Data in Mental Health Research (arxiv.org)

FW-NKF: Frequency-Weighted Neural Kalman Filters (arxiv.org)

Who Annotates in NLP? A Large-scale Assessment of Human Annotation Reporting between 2018 and 2025 (arxiv.org)

Quantitative Movement Testing: Measuring Patient Movements from a Single Smartphone Video (arxiv.org)

SeClaw: Spec-Driven Security Task Synthesis for Evaluating Autonomous Agents (arxiv.org)

Do Multimodal Agents Really Benefit from Tool Use? A Systematic Study of Capability Gains (arxiv.org)

SPADE-Bench: Evaluating Spontaneous Strategic Deception in Agents via Plan-Action Divergence (arxiv.org)

Monitoring Agentic Systems Before They're Reliable (arxiv.org)

AutoForest: Automatically Generating Forest Plots from Biomedical Studies with End-to-End Evidence Extraction and Synthesis (arxiv.org)

LeARN: Learnable and Adaptive Representations for Nonlinear Dynamics in System Identification (arxiv.org)

Efficient LLM Moderation with Multi-Layer Latent Prototypes (arxiv.org)

Introduction to Graph Neural Networks for Machine Learning Engineers (arxiv.org)

Efficient Weighted Sampling via Score-based Generative Models (arxiv.org)

Position: Neglecting the Sustainability of AI is Fuelling a Global AI Arms Race (arxiv.org)

TriLens: Per-Layer Logit-Lens Entropy for White-Box Hallucination Detection (arxiv.org)

Ideas in Inference-time Scaling can Benefit Generative Pre-training Algorithms (arxiv.org)

Interpretable Graph Kolmogorov-Arnold Networks for Multi-Cancer Classification and Biomarker Identification using Multi-Omics Data (arxiv.org)

Scalable Ride-Sourcing Vehicle Rebalancing with Service Accessibility Guarantee: A Constrained Mean-Field Reinforcement Learning Approach (arxiv.org)

GRANITE : a Byzantine-Resilient Dynamic Gossip Learning Framework (arxiv.org)

AdaCodec: A Predictive Visual Code for Video MLLMs (arxiv.org)

Evolutionary Discovery of Bivariate Bicycle Codes with LLM-Guided Search (arxiv.org)

Not All Errors Are Equal: A Systematic Study of Error Propagation in Large Language Model Inference (arxiv.org)

PaSBench-Video: A Streaming Video Benchmark for Proactive Safety Warning (arxiv.org)

Initialization is Half the Battle: Generating Diverse Images from a Guidance Potential Posterior (arxiv.org)

MASER: Modality-Adaptive Specialist Routing for Embodied 3D Spatial Intelligence (arxiv.org)

Learning When to Translate for Multilingual Reasoning (arxiv.org)

Ghost Tool Calls: Issue-Time Privacy for Speculative Agent Tools (arxiv.org)

Moment-Video: Diagnosing Temporal Fidelity of Video MLLMs on Momentary Visual Events (arxiv.org)

Why Not Hyperparameter-Friendly Optimisation? A Monotonic Adaptive Norm Rescaling Approach For Long-Tailed Recognition (arxiv.org)

Mitigating Perceptual Judgment Bias in Multimodal LLM-as-a-Judge via Perceptual Perturbation and Reward Modeling (arxiv.org)

Byte Pair Encoding for Efficient Time Series Forecasting (arxiv.org)

Beyond Discreteness: Sample Complexity Analysis of Straight-Through Estimator for 1-bit Quantization (arxiv.org)

Cooperation of Experts: Fusing Heterogeneous Information with Large Margin (arxiv.org)

Value-Free Policy Optimization via Reward Partitioning (arxiv.org)

FedCF: Fair Federated Conformal Prediction (arxiv.org)