Asap7772 / understanding-rlhfLinks
Learning from preferences is a common paradigm for fine-tuning language models. Yet, many algorithmic design decisions come into play. Our new work finds that approaches employing on-policy sampling or negative gradients outperform offline, maximum likelihood objectives.
☆31Updated last year
Alternatives and similar repositories for understanding-rlhf
Users that are interested in understanding-rlhf are comparing it to the libraries listed below
Sorting:
- This is code for most of the experiments in the paper Understanding the Effects of RLHF on LLM Generalisation and Diversity☆45Updated last year
- NeurIPS 2024 tutorial on LLM Inference☆47Updated 9 months ago
- Directional Preference Alignment☆59Updated 11 months ago
- Large language models (LLMs) made easy, EasyLM is a one stop solution for pre-training, finetuning, evaluating and serving LLMs in JAX/Fl…☆75Updated last year
- RL algorithm: Advantage induced policy alignment☆65Updated 2 years ago
- ☆150Updated 9 months ago
- ☆99Updated last year
- Code and Configs for Asynchronous RLHF: Faster and More Efficient RL for Language Models☆62Updated 4 months ago
- Self-Supervised Alignment with Mutual Information☆21Updated last year
- Official implementation of Bootstrapping Language Models via DPO Implicit Rewards☆44Updated 5 months ago
- Reinforcing General Reasoning without Verifiers☆81Updated 2 months ago
- ☆100Updated last year
- The repository contains code for Adaptive Data Optimization☆25Updated 9 months ago
- Dateset Reset Policy Optimization☆30Updated last year
- official implementation of ICLR'2025 paper: Rethinking Bradley-Terry Models in Preference-based Reward Modeling: Foundations, Theory, and…☆66Updated 5 months ago
- Advantage Leftover Lunch Reinforcement Learning (A-LoL RL): Improving Language Models with Advantage-based Offline Policy Gradients☆26Updated last year
- ICML 2024 - Official Repository for EXO: Towards Efficient Exact Optimization of Language Model Alignment☆58Updated last year
- Code for Paper (Policy Optimization in RLHF: The Impact of Out-of-preference Data)☆28Updated last year
- Code for the paper "VinePPO: Unlocking RL Potential For LLM Reasoning Through Refined Credit Assignment"☆169Updated 3 months ago
- ☆13Updated last month
- A Large-Scale, High-Quality Math Dataset for Reinforcement Learning in Language Models☆63Updated 6 months ago
- ☆100Updated last year
- Code for the arXiv preprint "The Unreasonable Effectiveness of Easy Training Data"☆48Updated last year
- Natural Language Reinforcement Learning☆96Updated last month
- ☆34Updated 8 months ago
- official implementation of paper "Process Reward Model with Q-value Rankings"☆61Updated 7 months ago
- CodeUltraFeedback: aligning large language models to coding preferences (TOSEM 2025)☆72Updated last year
- The official implementation of Self-Exploring Language Models (SELM)☆64Updated last year
- Rewarded soups official implementation☆60Updated last year
- ☆32Updated 10 months ago