reinforcement-learning-from-human-feedback

An Easy-to-use, Scalable and High-performance RLHF Framework (70B+ PPO Full Tuning & Iterative DPO & LoRA & RingAttention & RFT)

Safe RLHF: Constrained Value Alignment via Safe Reinforcement Learning from Human Feedback

A simulation framework for RLHF and alternatives. Develop your RLHF method without collecting human data.

Super-Efficient RLHF Training of LLMs with Parameter Reallocation

Okapi: Instruction-tuned Large Language Models in Multiple Languages with Reinforcement Learning from Human Feedback

A repo for RLHF training and BoN over LLMs, with support for reward model ensembles.

A Survey of Direct Preference Optimization (DPO)

Official code for ICML 2024 paper, "RIME: Robust Preference-based Reinforcement Learning with Noisy Preferences" (ICML 2024 Spotlight)

annotated tutorial of the huggingface TRL repo for reinforcement learning from human feedback connecting equations from PPO and GAE to the lines of code in the pytorch implementation

Shaping Language Models with Cognitive Insights

RLHF-Blender: A Configurable Interactive Interface for Learning from Diverse Human Feedback

[TSMC] Ask-AC: An Initiative Advisor-in-the-Loop Actor-Critic Framework

[AAMAS 2025] Privacy-preserving and Personalized RLHF, with convergence guarantees. The Code contains experiments for training multiple instances of GPT-2 for personalized sentiment aligned text generation.

Acceleration framework for Human Alignment Learning

This repository contains the implementation of a Reinforcement Learning with Human Feedback (RLHF) system using custom datasets. The project utilizes the trlX library for training a preference model that integrates human feedback directly into the optimization of language models.

LMRax is a framework built on JAX to train transformers language models by reinforcement learning, along with the reward model training.

Comparing various RLHF methods

Code for Bachelor thesis, The Human Factor: Addressing Diversity in Reinforcement Learning from Human Feedback.

Summaries of papers related to the alignment problem in NLP

Unlocking the Power of Generative AI: In-Context Learning, Instruction Fine-Tuning and Reinforcement Learning Fine-Tuning.