3 papers
cs.CL2026
Reinforcement Learning for LLM Post-Training: A Survey
Zhichao Wang, Kiran Ramnath, Bin Bi +9
Large language models (LLMs) trained via pretraining and supervised fine-tuning (SFT) can still produce harmful and misaligned outputs, or struggle in domains like math and coding.…
cs.CL2026
UFT: Unifying Fine-Tuning of SFT and RLHF/DPO/UNA through a Generalized Implicit Reward Function
Zhichao Wang, Bin Bi, Zixu Zhu +6
By pretraining on trillions of tokens, an LLM gains the capability of text generation. However, to enhance its utility and reduce potential harm, SFT and alignment are applied sequ…
cs.LG2026
UNA: A Unified Supervised Framework for Efficient LLM Alignment Across Feedback Types
Zhichao Wang, Bin Bi, Can Huang +7
RL alignment methods, including RLHF and DPO, are primarily based on pairwise preference data. Although scalar or score-based feedback has been collected in some settings, it is ra…