AI News

Atlas: Optimizing Deployment of Compound AI Workflows on Heterogeneous Clusters (arxiv.org)

A Semantic Model of Genetic Evidence: A Step Toward Bridging the Basic-Science-Clinic Gap (arxiv.org)

Towards Understanding Pause Token Fine-Tuning Dynamics: A Mode Retention Perspective (arxiv.org)

Patterns of Priming in Production: Lexical, Semantic and Structural Alignment in Language Model Generation (arxiv.org)

Shared circuits predict whether LLMs generalize across formats in arithmetic reasoning (arxiv.org)

A Roadmap for MEG Foundation Models (arxiv.org)

Dimension-Adaptive Batched Lipschitz Narrowing Without Knowing the Zooming Dimension (arxiv.org)

When Load-Balancing Goes Too Far: Expert Pruning in Over-Dispersed Mixture-of-Experts Models (arxiv.org)

GRACE: Graph-Grounded Reflective Agent Copilot Engine for Expert-in-the-Loop Knowledge Expansion (arxiv.org)

REFINE: LLM Refinement over Budgeted Text-Attributed Graphs for Personalized Medical Concept Representation (arxiv.org)

A Systematic Evaluation of Cross-Lingual Consistency Enhancement Methods in Multilingual Language Models (arxiv.org)

You Really Didn't Get That? Benchmarking Social Pragmatic Inference for Indirect and Playful Chinese Online Comments (arxiv.org)

Tracing Audio Grounding and Answer Selection in Audio LLMs (arxiv.org)

What Moves? Localized Motion Representations for Compositional Scene Control (arxiv.org)

When Quantization Breaks Memory: Recurrent-State Write-Back in Low-Precision Temporal Inference (arxiv.org)

PerfReasoning: How Well Do LLMs Reason on Hardware Performance? (arxiv.org)

HarvestBench: Measuring Whether LLM Agents Will Pay to Avoid Killing Animals (arxiv.org)

Corporate Language Model (CLM): Transforming Tacit and Fragmented Enterprise Knowledge into a Sovereign, Auditable, and Executable Corporate Intelligence Layer (arxiv.org)

Why Better Models Can Create Riskier Systems: Evidence from LLM Agents in Financial Markets (arxiv.org)

PetQA: Benchmarking Veterinary Knowledge and Clinical Reasoning (arxiv.org)

Hyperedge Anomaly Detection with Hypergraph Neural Network (arxiv.org)

Single-Query Black-Box Calibration Auditing via Logit Bias (arxiv.org)

A Comparative Study of Counterfactual Explainers for Graph Neural Networks Enabling Multiple Types of Graph Edit (arxiv.org)

NEAT-POCKET: Pocket-Conditioned Autoregressive 3D Molecular Generation with a Neighborhood-Guided Set Transformer (arxiv.org)

Enhancing Affine Maximizer Auctions with Correlation-Aware Payment (arxiv.org)

TeleTables: A Benchmark for Large Language Models in Telecom Table Interpretation (arxiv.org)

Constrained Sensing and Reliable State Estimation with Shallow Recurrent Decoders on a TRIGA Mark II Reactor (arxiv.org)

Graph Foundation Models for Recommendation: A Comprehensive Survey (arxiv.org)

Fractal basins trap latent reasoning (arxiv.org)

Hakken: Predicting future discoveries to fill the gaps in today's knowledge (arxiv.org)

On the Asymptotic Inadmissibility of Double Machine Learning Estimators Under Structure-Agnostic Models (arxiv.org)

Multi-Step Tool-Calling over Korean Open Public APIs: A Benchmark and a Data-Synthesis Recipe (arxiv.org)

High-dimensional censored MIDAS logistic regression for corporate survival forecasting (arxiv.org)

A Gaussian process model for chemoinformatics with application to the hazard classification of organic solvents (arxiv.org)

Learning from VAE Errors to support ECG-based Differential Diagnosis of Myocardial Scar (arxiv.org)

Confounding-Valid Conformal Inference for Counterfactual KPIs in Wireless Networks (arxiv.org)

Reconciling Universal and Uniform Learning with $Q$-Aggregation (arxiv.org)

Faster Learning under Relaxed Local Differential Privacy (arxiv.org)

Interpretability for Turing Machines (arxiv.org)

Optimal Stratified Allocation for Rare-Event Onset Forecasting in Dependent Sequences (arxiv.org)

A Constraint-Aware Generative Framework for Synthetic Origin-Destination Demand in Logistics Networks (arxiv.org)

FluxDisco: Symbolic Regression for Stoichiometric Dynamical Systems via Monte Carlo Graph Search (arxiv.org)

An Analysis of Self-supervised Pre-training with Dependent Samples (arxiv.org)

Minimax Lower Bound for Estimating Diffusion-based Local Intrinsic Dimension (arxiv.org)

Towards AI-Driven Nanomedicine Discovery: A Benchmark and Multimodal Learning Framework for Nano Self-Assembly Prediction (arxiv.org)

Self-Supervised Pretraining of Molecular Graph Encoders with LeJEPA (arxiv.org)

PAC-Bayesian Reconstruction Guarantees for Time Series Variational Autoencoders (arxiv.org)

Evaluating Large Language Models for Forced Outage Risk Prediction: Benefits and Comparison to Machine Learning (arxiv.org)

DODR: Deterministic Operator-Driven Reasoning in Latent Space (arxiv.org)

Fast Gauss Sums via Flash Attention (arxiv.org)