allenai

/

Llama-3.1-Tulu-3-70B

like 44

Follow

Ai2 1.78k

Hello, I noticed a spelling error in the paper. In Section 5.4 "Preference Tuning Recipe and Analyses," under the subsection "Comparison Between PPO and DPO," the sentence:

"Furthermore, iterating with RM and PPO is more expensive then iterating with DPO"
should use "than" instead of "then." This is likely a typo, as "more ... than" is the correct comparative structure in English.

Thank you for your attention to this!