Papers with scalable preference data construction strategy

1 papers
Finding the Sweet Spot: Preference Data Construction for Scaling Preference Optimization (2025.acl-long)

Copied to clipboard

Challenge: Large language models generate unintended outputs due to their unsupervised nature.
Approach: They propose a method to construct preference pairs of selected and rejected LLMs by repeated random sampling to improve alignment performance.
Outcome: The proposed method improves performance as the sample size increases.

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