RLHFlow / Online-RLHFLinks
A recipe for online RLHF and online iterative DPO.
☆520Updated 5 months ago
Alternatives and similar repositories for Online-RLHF
Users that are interested in Online-RLHF are comparing it to the libraries listed below
Sorting:
- Recipes to train reward model for RLHF.☆1,380Updated 2 months ago
- RewardBench: the first evaluation tool for reward models.☆604Updated last week
- ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Search (NeurIPS 2024)☆639Updated 5 months ago
- ☆208Updated last month
- ☆300Updated 3 weeks ago
- This is the repository that contains the source code for the Self-Evaluation Guided MCTS for online DPO.☆318Updated 10 months ago
- The official implementation of Self-Play Preference Optimization (SPPO)☆566Updated 5 months ago
- Codebase for Iterative DPO Using Rule-based Rewards☆247Updated 2 months ago
- (ICML 2024) Alphazero-like Tree-Search can guide large language model decoding and training☆276Updated last year
- ☆203Updated 4 months ago
- [NeurIPS 2024] BAdam: A Memory Efficient Full Parameter Optimization Method for Large Language Models☆259Updated 3 months ago
- [ICML 2024] LESS: Selecting Influential Data for Targeted Instruction Tuning☆456Updated 8 months ago
- ☆331Updated 2 weeks ago
- [NeurIPS 2024] SimPO: Simple Preference Optimization with a Reference-Free Reward☆897Updated 4 months ago
- A simple toolkit for benchmarking LLMs on mathematical reasoning tasks. 🧮✨☆226Updated last year
- A series of technical report on Slow Thinking with LLM☆699Updated 2 weeks ago
- RLHF implementation details of OAI's 2019 codebase☆187Updated last year
- A library with extensible implementations of DPO, KTO, PPO, ORPO, and other human-aware loss functions (HALOs).☆857Updated 2 weeks ago
- Implementation for "Step-DPO: Step-wise Preference Optimization for Long-chain Reasoning of LLMs"☆369Updated 5 months ago
- A lightweight reproduction of DeepSeek-R1-Zero with indepth analysis of self-reflection behavior.☆241Updated 2 months ago
- ☆782Updated last month
- A large-scale, fine-grained, diverse preference dataset (and models).☆342Updated last year
- ☆317Updated 9 months ago
- verl-agent is an extension of veRL, designed for training LLM/VLM agents via RL. verl-agent is also the official code for paper "Group-in…☆398Updated this week
- ☆288Updated 11 months ago
- L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning☆222Updated last month
- Implementation of paper Data Engineering for Scaling Language Models to 128K Context☆463Updated last year
- MetaMath: Bootstrap Your Own Mathematical Questions for Large Language Models☆436Updated last year
- Code for Paper (ReMax: A Simple, Efficient and Effective Reinforcement Learning Method for Aligning Large Language Models)☆185Updated last year
- ☆276Updated 5 months ago