Projects sharing features with PRIME

🌠 A PO st-training recipe for scaling R L on A dvanced R eason I ng model S 🚀

2025/10/23 🔥🔥Our paper is accepted by NeurIPS 2025.🔥🔥 - 2025/04/22 Released our Paper on arXiv. See here - 2025/03/24 We re-implement our algorithm based on verl. ✨✨ Key features: (1) add ~50 additional metrics to comprehensively monitor the training process and stability, (2) add a…

This repository contains the code for our paper titled "Optimizing Test-Time Compute via Meta Reinforcement Finetuning." In this work, we introduce a novel approach to optimizing test-time compute through meta reinforcement learning, aiming to balance the efficiency and discovery capabilities of…

Enhancing Vision-Language Model Training with Reinforcement Learning in Synthetic Worlds for Real-World Success

🔧✨Tool-Star: Empowering Multi-Tool Collaborative Web Agent via Reinforcement Learning

Princeton University \& PKU \& UIUC \& University of Chicago \& ByteDance Seed

📄 Paper Link 🤗 VerIPO-7B-v1.0

December 5, 2025 🔍 We propose entropy ratio clipping​ (ERC) to impose a global constraint on the output distribution of the policy model. Experiments demonstrate that ERC can significantly improve the stability of off-policy training. 📄 The paper is available on arXiv.

25/08/20 : Aceept as EMNLP 2025 Main Conference paper

Free Process Rewards without Process Labels

Paper - Abstract - Updates - Quick Start - Installation - Download the datasets - Create Experiment Script - Single GPU Training (Only for Rho models) - Running the experiments - Code Structure - Initial SFT Checkpoints - Acknowledgement - Citation

💜 Confucius Demo | 🤗 Hugging Face | 🤖 ModelScope | ⌨️ GitHub | 📚 Paper | 💬 Wechat

An Open Source Approach to Scaling Up Reinforcement Learning on the Base Model

This repository accompanies the paper Let's Verify Step by Step and presents the PRM800K dataset introduced there. PRM800K is a process supervision dataset containing 800,000 step-level correctness labels for model-generated solutions to problems from the MATH dataset. More information on…

Free Process Rewards without Process Labels

rllm is an asynchronous reinforcement learning framework for training language agents. It provides a unified pipeline that runs the same agent code for both evaluation and training, automatically capturing traces for gradient computation. The framework supports distributed reinforcement learning across multiple GPUs and nodes using pluggable backends, and executes agents in isolated sandboxes—either locally or in the cloud—for safe and scalable rollout collection. It trains agents built with LangGraph, SmolAgents, OpenAI Agents SDK, or custom frameworks without requiring core logic changes. T

This repository contains the official implementation of AutoPSV: Automated Process-Supervised Verifier, accepted at NeurIPS 2024 (poster).

👋 Hi, everyone! verl is a RL training library initiated by ByteDance Seed team and maintained by the verl community.

G2RPO-A: Guided Group Relative Policy Optimization with Adaptive Guidance

DARS: Dynamic Action Re-Sampling to Enhance Coding Agent Performance by Adaptive Tree Traversal

The official repository for "Parallel-R1: Towards Parallel Thinking via Reinforcement Learning".

🚀 Segment Policy Optimization: Effective Segment-Level Credit Assignment in RL for Large Language Models 🌟

StepSearch: Igniting LLMs Search Ability via Step-Wise Proximal Policy Optimization