Showing 2024Show all
2 papers · 1 filter
cs.LG2024
Towards Improved Preference Optimization Pipeline: from Data Generation to Budget-Controlled Regularization
Zhuotong Chen, Fang Liu, Jennifer Zhu +2
Direct Preference Optimization (DPO) and its variants have become the de facto standards for aligning large language models (LLMs) with human preferences or specific goals. However…
cs.CL2024
DFlow: Diverse Dialogue Flow Simulation with Large Language Models
Wanyu Du, Song Feng, James Gung +4
Developing language model-based dialogue agents requires effective data to train models that can follow specific task logic. However, most existing data simulation methods focus on…