WeiXiongUST/Building-Math-Agents-with-Multi-Turn-Iterative-Preference-Learning
This is an official implementation of the paper ``Building Math Agents with Multi-Turn Iterative Preference Learning'' with multi-turn DPO and KTO.
Ph.D. Student in computer science at UIUC; machine learning theory and RLHF.
Repositories
WeiXiongUST/Decentralized-Proximal-Algorithm-with-Variance-Reduction
This is the code used for the paper "PMGT-VR: A decentralized proximal-gradient algorithmic framework with variance reduction", prepint.
WeiXiongUST/WeiXiongUST
WeiXiongUST/NeMo-Skills
A pipeline to improve skills of large language models
WeiXiongUST/Reinforce-Ada
An adaptive sampling framework for Reinforce-style LLM post training.
WeiXiongUST/Minimal-RL
WeiXiongUST/lora_rm
WeiXiongUST/code_trn
WeiXiongUST/CS510-SP25-Course-Project
WeiXiongUST/Online-RLHF
WeiXiongUST/math_eval
Codebase for Iterative DPO Using Rule-based Rewards
WeiXiongUST/Self-rewarding-reasoning-LLM
Recipes to train the self-rewarding reasoning LLMs.
WeiXiongUST/self-rw
WeiXiongUST/eval_math
WeiXiongUST/rm_dev
WeiXiongUST/raft_verl
WeiXiongUST/BanditLib
Library of contextual bandits algorithms
WeiXiongUST/infer_math
WeiXiongUST/OpenRLHF
An Easy-to-use, Scalable and High-performance RLHF Framework (70B+ PPO Full Tuning & Iterative DPO & LoRA & RingAttention & RFT)
WeiXiongUST/dpo_math
WeiXiongUST/RLHF-Reward-Modeling-dev
Recipes to train reward model for RLHF.
WeiXiongUST/feedbackagent
WeiXiongUST/axolotl
Go ahead and axolotl questions
WeiXiongUST/CS598
WeiXiongUST/RAFT
This is an official implementation of the Reward rAnked Fine-Tuning Algorithm (RAFT), also known as iterative best-of-n fine-tuning or rejection sampling fine-tuning.
WeiXiongUST/alignment-handbook
Robust recipes to align language models with human and AI preferences
WeiXiongUST/ai-for-grant-writing
A curated list of resources for using LLMs to develop more competitive grant applications.
WeiXiongUST/iterative-rlhf
WeiXiongUST/ToRA
ToRA is a series of Tool-integrated Reasoning LLM Agents designed to solve challenging mathematical reasoning problems by interacting with tools [ICLR'24].