AI Research Papers Daily

Unified Multimodal Chain-of-Thought Reward Model through Reinforcement Fine-Tuning

Absolute Zero: Reinforced Self-play Reasoning with Zero Data

RADLADS: Rapid Attention Distillation to Linear Attention Decoders at Scale

FlexiAct: Towards Flexible Action Control in Heterogeneous Scenarios

RetroInfer: A Vector-Storage Approach for Scalable Long-Context LLM Inference

An Empirical Study of Qwen3 Quantization

Decoding Open-Ended Information Seeking Goals from Eye Movements in Reading

Multi-Agent System for Comprehensive Soccer Understanding

Geospatial Mechanistic Interpretability of Large Language Models

HoloTime: Taming Video Diffusion Models for Panoramic 4D Scene Generation

VITA-Audio: Fast Interleaved Cross-Modal Token Generation for Efficient Large Speech-Language Model

InfoVids: Reimagining the Viewer Experience with Alternative Visualization-Presenter Relationships

SWE-smith: Scaling Data for Software Engineering Agents

Teaching Models to Understand (but not Generate) High-risk Data

Scenethesis: A Language and Vision Agentic Framework for 3D Scene Generation

Invoke Interfaces Only When Needed: Adaptive Invocation for Large Language Models in Question Answering

Auto-SLURP: A Benchmark Dataset for Evaluating Multi-Agent Frameworks in Smart Personal Assistant

Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systems

Alpha Excel Benchmark