Search Results for author: Dongyoung Go

Found 2 papers, 2 papers with code

Compositional preference models for aligning LMs

1 code implementation17 Oct 2023 Dongyoung Go, Tomasz Korbak, Germán Kruszewski, Jos Rozen, Marc Dymetman

As language models (LMs) become more capable, it is increasingly important to align them with human preferences.

Aligning Language Models with Preferences through f-divergence Minimization

1 code implementation16 Feb 2023 Dongyoung Go, Tomasz Korbak, Germán Kruszewski, Jos Rozen, Nahyeon Ryu, Marc Dymetman

We show that Jensen-Shannon divergence strikes a good balance between these objectives, and frequently outperforms forward KL divergence by a wide margin, leading to significant improvements over prior work.

Cannot find the paper you are looking for? You can Submit a new open access paper.