docs: Add XPO (2405.21046) to Paper Index #5068
Merged
+15
−0
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- Add Exploratory Preference Optimization entry under Online DPO section - Note that paper defines α > 0 and β > 0 but does not specify numerical values - Config uses TRL defaults (alpha=1e-5, beta=0.1)
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary
Adds the XPO (Exploratory Preference Optimization) paper entry to the paper index under the Online Direct Preference Optimization section.
Relates to #4407
Note on hyperparameters: The XPO paper (arXiv 2405.21046) defines α > 0 (optimism coefficient) and β > 0 (KL regularization) in Algorithm 1 but does not specify numerical values in the paper — the experimental details are not publicly accessible, and the paper authors did not release a standalone codebase. The configuration uses TRL defaults (
alpha=1e-5,beta=0.1) and this is clearly noted in the entry.