Eliminating Biased Length Reliance of Direct Preference Optimization via Down-Sampled KL Divergence
-
jiazhengli/Pythia-2.8B-HH-RLHF-Iterative-SamPO
Text Generation • Updated • 11 -
jiazhengli/Pythia-2.8B-TLDR-Iterative-SamPO
Text Generation • Updated • 13 -
Junrulu/Llama-3-8B-Instruct-Iterative-SamPO
Text Generation • Updated • 20 -
Eliminating Biased Length Reliance of Direct Preference Optimization via Down-Sampled KL Divergence
Paper • 2406.10957 • Published • 1