Stars
Official Implementation of the Paper "Let's Predict Step by Step"
[ACL 2025 Main] MMBoundary: Advancing MLLM Knowledge Boundary Awareness through Reasoning Step Confidence Calibration
A Framework for LLM-based Multi-Agent Reinforced Training and Inference
This is the code repository of the video reasoning benchmark MMR-V
Video-Holmes: Can MLLM Think Like Holmes for Complex Video Reasoning?
Train your Agent model via our easy and efficient framework
Official implementation of the paper "Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Space"
This is the repository of DEER, a Dynamic Early Exit in Reasoning method for Large Reasoning Language Models.
Visual Planning: Let's Think Only with Images
Parallel Scaling Law for Language Model — Beyond Parameter and Inference Time Scaling
ACL'2025: SoftCoT: Soft Chain-of-Thought for Efficient Reasoning with LLMs. and preprint: SoftCoT++: Test-Time Scaling with Soft Chain-of-Thought Reasoning
OpenThinkIMG is an end-to-end open-source framework that empowers LVLMs to think with images.
Structural Entropy Guided Agent for Detecting and Repairing Knowledge Deficiencies in LLMs
VeOmni: Scaling any Modality Model Training to any Accelerators with PyTorch native Training Framework
DeerFlow is a community-driven Deep Research framework, combining language models with tools like web search, crawling, and Python execution, while contributing back to the open-source community.
Seed-Coder is a family of lightweight open-source code LLMs comprising base, instruct and reasoning models, developed by ByteDance Seed.
EchoInk-R1: Exploring Audio-Visual Reasoning in Multimodal LLMs via Reinforcement Learning [🔥The Exploration of R1 for General Audio-Visual Reasoning with Qwen2.5-Omni]
Evaluation Code Repo for Paper "PolyMath: Evaluating Mathematical Reasoning in Multilingual Contexts"
GPG: A Simple and Strong Reinforcement Learning Baseline for Model Reasoning
NoisyRollout: Reinforcing Visual Reasoning with Data Augmentation
A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning
Unleashing the Power of Reinforcement Learning for Math and Code Reasoners