AI News

PACEvolve: Enabling Progress-Aware Consistent Evolution (arxiv.org)

AnyView: Synthesizing Any Novel View in Dynamic Scenes (arxiv.org)

Code-to-Harness: Distilling Black-Box Optimizers from Self-Play (arxiv.org)

ProactiveBench: Can Streaming Video Models Really Interact Like Humans? (arxiv.org)

Distortion of AI Alignment Revisited: RLHF is a Decent Utilitarian Aligner (arxiv.org)

Explaining Time Series Forecasting with Horizon-Resolved Attribution (arxiv.org)

Geometric-to-Semantic Spherical Transfer Learning for Cortical Sulci Labeling (arxiv.org)

Correlation-Guided Fast Machine Unlearning via Hessian Analysis (arxiv.org)

SIMS: Scale-Invariant Merit-Function-Based Scalarization for Multi-Task Learning (arxiv.org)

Poisson-Corrector Complexity Bounds for Moreau--Yosida Unadjusted Langevin Sampling (arxiv.org)

Clustering-Based Balanced Sampling and Allocation with Data Parallelism for High-Performance Fine-Tuning (arxiv.org)

SCOPE-OPSD: Fisher-Conditioned Privileged Subspaces for On-Policy Self-Distillation (arxiv.org)

TokenMapper: A Step Toward Interoperable Speech Token Translation (arxiv.org)

Quality-Constrained Routing over a Fixed Pool of Quantized Mixture-of-Experts Instances (arxiv.org)

Where Decoder Cosine Similarity Fails for SAE Feature Flow Discovery (arxiv.org)

$\text{GSF-}\chi$: Global Stereochemical Fields for Chiral Graph Transformers (arxiv.org)

Temporal Recurrence Favors Fewer Layers (arxiv.org)

Split Conformal Prediction with Label-Shift-Adjusted Bayesian Scores (arxiv.org)

Certified AI Triage of ICU Alarms (arxiv.org)

LatentVerse: A Framework for Understanding Shared and Modality-Specific Information in Multimodal Latent Representations (arxiv.org)

When Connected Does Not Mean Similar: Charting the Homophily Boundary of SNAP-KG for Streaming Entity Integration (arxiv.org)

ParaRecover: A Process-Level Benchmark for Error Localization and Recovery in Parallel Tool-Use Agents (arxiv.org)

Theoretical Guarantees for One-Shot Magnitude Pruning and Compute-Adaptive Early Exit (arxiv.org)

Simulating Disengaged Students to Evaluate LLM-based Tutors (arxiv.org)

Sampling via Decision-Flow: Training-Free Extraction of Improved Latent Reasoning Paths in Large Language Models (arxiv.org)

A Differentially Private Federated Proximal Optimization Framework for Customer Churn Prediction in Heterogeneous Federated Telecom Networks (arxiv.org)

SAGE-Loop: Reliable Closed-Loop LLM-Driven AutoML with Trial-and-Correction and Adaptive Ensembling (arxiv.org)

Observation-Anchored Selective Assimilation for Longitudinal Tumor-State Proxy Forecasting in Post-Treatment Glioma (arxiv.org)

Granularity-Adaptive Credit Assignment for Long-Horizon LLM Agent Reinforcement Learning (arxiv.org)

MInTRL: Off-policy Intervention can boost On-policy RL (arxiv.org)

RiPPLE: Cross-Space Performance Prediction from Early Training for Neural Architecture Search (arxiv.org)

Amortized Low-Rank Adaptation for Model-Based Reinforcement Learning (arxiv.org)

Reinforcement Learning over Patient Trajectories for Clinical Reasoning in EHR Foundation Models (arxiv.org)

Adaptive Chemotherapy Control under Tumor Heterogeneity via Reinforcement Learning (arxiv.org)

The Rank the Task Demands: A Causal Rank Law for Matrix Memories Trained on Group Composition (arxiv.org)

CRFCAN: A Complex-Valued Cross-Domain Residual Network for Joint Channel and Phase Noise Estimation in Sub-THz OFDM Systems (arxiv.org)

PLSP (Pre-hoc Liminal Space Profiling): OOD Prediction over Detection -- An Anticipatory Approach for Machine Learning Model Reliability (arxiv.org)

Patient-Reported Survey Data Improve Prediction of Opioid Use Disorder (arxiv.org)

Predicting Collision Cross Sections with GRACE: Geometric Residual Adduct Conditioning via Early-fusion (arxiv.org)

Certifying Concept Unlearning in Text-to-Image Diffusion Models (arxiv.org)

GUIDE: Generative Utility Inference and Decision Engine (arxiv.org)

Almost Sure Convergence Analysis of Stochastic Gradient Methods with Clipping and Additive Noise (arxiv.org)

Scalable Discrete-to-Continuous Channel Simulation for Compression and Privacy (arxiv.org)

Reinforcement Learning for Syndrome Extraction (arxiv.org)

Toward Reliable Railway-Bogie Response Prediction Using Multifidelity TDNN and Physics-Informed Residual Learning (arxiv.org)

Inverting Self-Triggered Control: Adversarial Reinforcement Learning for Sparse Denial-of-Service Attacks (arxiv.org)

Certified Safety Curation: Distribution-Free Guarantees for Safe Offline Reinforcement Learning (arxiv.org)

QTrans: A Quantum Transformer for Sentiment Classification (arxiv.org)

Can We Trust LLM Judges: A Study of Capability-Dependent Biases and Multi-Judge Ensemble for Bias Calibration (arxiv.org)

Fixed State, Long Reach: What a Constant-Size Cache Buys Block Diffusion at Scale (arxiv.org)