Earlier Tokens Contribute More: Learning Direct Preference Optimization From Temporal Decay Perspective - researchr publication authors

researchr

You are not signed in
Sign in
Sign up

Ruichen Shao, Bei Li, Gangao Liu, Yang Chen, Zhouxiang, Jingang Wang, Xunliang Cai, Peng Li. Earlier Tokens Contribute More: Learning Direct Preference Optimization From Temporal Decay Perspective. In The Thirteenth International Conference on Learning Representations, ICLR 2025, Singapore, April 24-28, 2025. OpenReview.net, 2025. [doi]

This author has not been identified. Look up 'Ruichen Shao' in GoogleThis author has not been identified. Look up 'Bei Li' in GoogleThis author has not been identified. Look up 'Gangao Liu' in GoogleThis author has not been identified. Look up 'Yang Chen' in GoogleThis author has not been identified. Look up 'Zhouxiang' in GoogleThis author has not been identified. Look up 'Jingang Wang' in GoogleThis author has not been identified. Look up 'Xunliang Cai' in GoogleThis author has not been identified. Look up 'Peng Li' in Google

runs on WebDSL