Skip to main content

🏢 Westlake University

Direct Preference Optimization Using Sparse Feature-Level Constraints
·2078 words·10 mins
AI Generated 🤗 Daily Papers Natural Language Processing Large Language Models 🏢 Westlake University
Feature-level constrained Preference Optimization (FPO) boosts LLM alignment efficiency and stability by using sparse autoencoders and feature-level constraints, achieving significant improvements ove…