AI研究论文每日精选

LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers

SurveyX: Academic Survey Automation via Large Language Models

Mol-LLaMA: Towards General Understanding of Molecules in Large Molecular Language Model

PhotoDoodle: Learning Artistic Image Editing from Few-Shot Pairwise Data

MaskGWM: A Generalizable Driving World Model with Video Mask Reconstruction

SIFT: Grounding LLM Reasoning in Contexts via Stickers

VLM^2-Bench: A Closer Look at How Well VLMs Implicitly Link Explicit Matching Visual Cues

LightThinker: Thinking Step-by-Step Compression

Is Safety Standard Same for Everyone? User-Specific Safety Evaluation of Large Language Models

MoBA: Mixture of Block Attention for Long-Context LLMs

StructFlowBench: A Structured Flow Benchmark for Multi-turn Instruction Following

Towards Fully-Automated Materials Discovery via Large-Scale Synthesis Dataset and Expert-Level LLM-as-a-Judge

Evaluating Multimodal Generative AI with Korean Educational Standards

MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models

Think Inside the JSON: Reinforcement Strategy for Strict LLM Schema Adherence

The Relationship Between Reasoning and Performance in Large Language Models -- o3 (mini) Thinks Harder, Not Longer

ReQFlow: Rectified Quaternion Flow for Efficient and High-Quality Protein Backbone Generation

FantasyID: Face Knowledge Enhanced ID-Preserving Video Generation

One-step Diffusion Models with f-Divergence Distribution Matching

InterFeedback: Unveiling Interactive Intelligence of Large Multimodal Models via Human Feedback

KITAB-Bench: A Comprehensive Multi-Domain Benchmark for Arabic OCR and Document Understanding

EgoSpeak: Learning When to Speak for Egocentric Conversational Agents in the Wild

Superintelligent Agents Pose Catastrophic Risks: Can Scientist AI Offer a Safer Path?

Tree-of-Debate: Multi-Persona Debate Trees Elicit Critical Thinking for Scientific Comparative Analysis

mStyleDistance: Multilingual Style Embeddings and their Evaluation

CrossOver: 3D Scene Cross-Modal Alignment

PLDR-LLMs Learn A Generalizable Tensor Operator That Can Replace Its Own Deep Neural Net At Inference

WHAC: World-grounded Humans and Cameras

Rare Disease Differential Diagnosis with Large Language Models at Scale: From Abdominal Actinomycosis to Wilson's Disease

Benchmarking LLMs for Political Science: A United Nations Perspective

Learning to Discover Regulatory Elements for Gene Expression Prediction

UPCORE: Utility-Preserving Coreset Selection for Balanced Unlearning

JL1-CD: A New Benchmark for Remote Sensing Change Detection and a Robust Multi-Teacher Knowledge Distillation Framework

Beyond No: Quantifying AI Over-Refusal and Emotional Attachment Boundaries