@tengx6 : ❓ Can we align large language models without hyperparameters? The answer is SimPER! 🚀 Excited to share our ICLR paper: SimPER: A Minimalist Approach to Preference Alignment without Hyperparameters! 🎯 SimPER eliminates the need for complex tuning and reference models, • TwiCopy

Teng Xiao

@tengx6

+ Follow

PhD student at @penn_state. Machine Learning and Reinforcement Learning

ID: 1171425990649634820

linkhttps://tengxiao1.github.io/ calendar_today10-09-2019 14:11:42

39 Tweet

87 Takipçi

590 Takip Edilen

Teng Xiao

@tengx6

6 months ago

❓ Can we align large language models without hyperparameters? The answer is SimPER! 🚀 Excited to share our ICLR paper: SimPER: A Minimalist Approach to Preference Alignment without Hyperparameters! 🎯 SimPER eliminates the need for complex tuning and reference models,

thumb_up_off_alt8

chat_bubble_outline1

repeat3

shareShare