Papers by Wangjie Gan

GenGO

Omakase Read Later

Search

Papers by Wangjie Gan

Findings

2026.findings-acl

Group Fine-Tuning
GFT
unified post-training framework
Group Advantage Learning
Dynamic Coefficient Rectification
SFT-based methods
RL training
entropy collapse
reward sparsity
optimization

2026

Language Models

1 papers

GFT: From Imitation to Reward Fine-Tuning with Unbiased Group Advantages and Dynamic Coefficient Rectification (2026.findings-acl)

Copied to clipboard

Wangjie Gan, Miao Pan, Linbo Xi, Wenqi Zhang, Jintao Chen, Jianwei Yin, Xuhong Zhang

Language Models

Challenge:	Existing studies have demonstrated that supervised fine-tuning and reinforcement learning are effective in integrating knowledge injection with robust generalization.
Approach:	They propose a unified post-training framework that addresses intrinsic limitations of supervised fine-tuning and reinforcement learning.
Outcome:	The proposed framework surpasses SFT-based methods and yields policies that integrate more smoothly with subsequent RL training.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations