reinforcement-learning-from-human-feedback

An Easy-to-use, Scalable and High-performance RLHF Framework based on Ray (PPO & GRPO & REINFORCE++ & vLLM & Ray & Dynamic Sampling & Async Agentic RL)

#数据仓库#Safe RLHF: Constrained Value Alignment via Safe Reinforcement Learning from Human Feedback

#自然语言处理#A simulation framework for RLHF and alternatives. Develop your RLHF method without collecting human data.

#大语言模型#Super-Efficient RLHF Training of LLMs with Parameter Reallocation

#自然语言处理#Okapi: Instruction-tuned Large Language Models in Multiple Languages with Reinforcement Learning from Human Feedback

#大语言模型#A Survey of Direct Preference Optimization (DPO)

CodeUltraFeedback: aligning large language models to coding preferences (TOSEM 2025)

#计算机科学#A repo for RLHF training and BoN over LLMs, with support for reward model ensembles.

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

#自然语言处理#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

[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 gener...

Acceleration framework for Human Alignment Learning

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

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 t...

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

#大语言模型#Comparing various RLHF methods

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

#自然语言处理#Summaries of papers related to the alignment problem in NLP