AI News

HeRE: A Novel Benchmark for Hebrew Relation Extraction via LLM-Guided Denoising of Knowledge Graph Alignments (arxiv.org)

Learning the Shared Latent Structure of Cryo-EM Inverse Problems (arxiv.org)

Towards Reconstructing Geographically Diverse Architecture with 3D Foundation Models (arxiv.org)

Beyond Latent Personalization: Integrating HRV-Conditioned Personalization via LLM for Interpretable Stress Detection (arxiv.org)

GVPO: Group Variance Policy Optimization for Large Language Model Post-Training (arxiv.org)

Rethinking the design space of Vanilla Pixel Diffusion Transformers (arxiv.org)

Behavioral Couplings: From Formation to Behavioral Emergence in Language Models (arxiv.org)

GLOBE: Evolving Structured Agentic Memory for E-commerce Product Optimization in Generative Search (arxiv.org)

Has GPT-6-Astra Achieved Embodied Intelligence? (arxiv.org)

Where Prefill Expert Reduction Hurts MoE Models: A Few Prompt Tokens Carry Most of the Short-Answer F1 Loss (arxiv.org)

Learnable Splines for Unstructured Neural Operators (arxiv.org)

Null Controls Are Necessary but Not Sufficient: Evaluating Acceptance Guards for Scientific Discovery and Interpretability (arxiv.org)

An Adaptive-Horizon Graph Neural Operator for coupled surrogate morphodynamic modelling (arxiv.org)

A Survey of Explainable Reinforcement Learning: Targets, Methods and Needs (arxiv.org)

Accelerated Execution of Bayesian Neural Networks using a Single Probabilistic Forward Pass and Code Generation (arxiv.org)

Closed-form Covariance Functions in Space and Time from B-spline Spectral Densities (arxiv.org)

RAW-Explainer: Post-hoc Explanations of Path Reasoning Graph Neural Networks for Knowledge Graph Completion (arxiv.org)

FRACTAL: A Self-Similar Multi-Agent Orchestration Architecture with Formal Communication, Convergence, and Depth Bounds (arxiv.org)

When is a large language model a good calibration agent? Physics-labeled prompting for budgeted engineering simulator calibration (arxiv.org)

Scalable Levenberg-Marquardt Training Framework for Deep Learning (arxiv.org)

A Feature Selection Layer Approach for Multi-Class Classification on Neural Networks (arxiv.org)

ARBOR: Locating Reasoning Failures in Long Chains of Thought through Reasoning State Tracking (arxiv.org)

NoveltyScope: A Multi-Agent Research Assistant for Evidence-Grounded Novelty Assessment of Scientific Papers (arxiv.org)

Explore Separately, Adopt Selectively: Execution-Guided Tool Use over Large Tool Libraries (arxiv.org)

Diverse Reasoning Paths Matter: Symbolic Equivalence Augmentation for Enhancing RL Exploration (arxiv.org)

Hard Instances on Demand (arxiv.org)

Empowering Bayes with sources of uncertainty: from confidence modelling to sparse estimation (arxiv.org)

LongForge: Measuring and Improving Long-Context Instruction Data via Evidence-Assertion Reasoning Graphs (arxiv.org)

Algorithm Modelization: Transforming Optimization Algorithms into Evolvable Intelligence (arxiv.org)

QuAD: Quantitative Assessment via Distribution-Alignment for Qualitative Data Quality Enabled by a Content-Aware Multi-Agent System (arxiv.org)

CIIR - Class-Indexed Interface Representation Learning (arxiv.org)

What Accuracy Hides In Learned Transformer Programs (arxiv.org)

Batched Contextual Reinforcement II: Does RLVR Need Long Rollout Windows? (arxiv.org)

Learning to Be Guided: Shaping Latent Geometry for Classifier-Free Guidance (arxiv.org)

Benchmarking and Enhancing Agentic Causal Discovery (arxiv.org)

2D Stability Selection: Design Jittering to Exclude Correlated Irrelevant Features (arxiv.org)

Partial-Revelation Panel Forecasting: Predictive Information Gain, Allocation, and Coarsening-Consistent Operators (arxiv.org)

Ori: Drift-Aware Streaming $k$-PCA for Very Low-Rank LLM Training (arxiv.org)

Next Token or Vector? Learning How Far to Decode (arxiv.org)

Agentic In-Context Learning: What Algorithms Can Transformers Learn in Closed-Loop Interaction? (arxiv.org)

On the tradeoff between robustness and geometric accuracy in adversarial training (arxiv.org)

LexPilot: Benchmarking the Proactive Interaction Ability of Legal LLMs with Laypeople in Chinese Civil Case Consultation (arxiv.org)

WebCoder: Scaling General Frontend Development with Artifact-Grounded Agentic Reinforcement Learning (arxiv.org)

Rethinking GRPO for Large Reasoning Model Safety Alignment (arxiv.org)

Reusing Calibration Gains across Model Representations (arxiv.org)

Where Does the Performance Go? Understanding and Recovering Performance Regressions in C-to-Rust Translation (arxiv.org)

Where Research Ideation Begins: Forward-Citation Supervision for Actionable Limitation Generation (arxiv.org)

ASE: Attribution-Guided Semantic Evolution for Time Series Forecasting (arxiv.org)

Causal Value Attribution of Marketing Labels in LLM-Generated Ads: A Deconfounder Approach (arxiv.org)

Acquisition Function-Guided Generation for Bayesian Optimization over Discrete Spaces (arxiv.org)