AI研究论文每日精选

Kanana: Compute-efficient Bilingual Language Models

GHOST 2.0: generative high-fidelity one shot transfer of heads

TheoremExplainAgent: Towards Multimodal Explanations for LLM Theorem Understanding

Towards an AI co-scientist

Plutus: Benchmarking Large Language Models in Low-Resource Greek Finance

Language Models' Factuality Depends on the Language of Inquiry

Can Large Language Models Detect Errors in Long Chain-of-Thought Reasoning?

Rank1: Test-Time Compute for Reranking in Information Retrieval

Agentic Reward Modeling: Integrating Human Preferences with Verifiable Correctness Signals for Reliable Reward Systems

Project Alexandria: Towards Freeing Scientific Knowledge from Copyright Burdens via LLMs

Can Language Models Falsify? Evaluating Algorithmic Reasoning with Counterexample Creation

Distill Any Depth: Distillation Creates a Stronger Monocular Depth Estimator

VEM: Environment-Free Exploration for Training GUI Agent with Value Environment Model

CritiQ: Mining Data Quality Criteria from Human Preferences

Drop-Upcycling: Training Sparse Mixture of Experts with Partial Re-initialization

BIG-Bench Extra Hard

Adapting Automatic Speech Recognition for Accented Air Traffic Control Communications

FSPO: Few-Shot Preference Optimization of Synthetic Preference Data in LLMs Elicits Effective Personalization to Real Users

AISafetyLab: A Comprehensive Framework for AI Safety Evaluation and Improvement

MolSpectra: Pre-training 3D Molecular Representation with Multi-modal Energy Spectra

Towards Optimal Multi-draft Speculative Decoding

MMKE-Bench: A Multimodal Editing Benchmark for Diverse Visual Knowledge

PosterSum: A Multimodal Benchmark for Scientific Poster Summarization

DOEI: Dual Optimization of Embedding Information for Attention-Enhanced Class Activation Maps