Update SatImp Hyperparameter and add ES retain metric.#192
Open
Puning97 wants to merge 4 commits into
Open
Conversation
We update the following components: Extraction Strength (ES) metric for retain data. Hyperparameter default setting for SatImp New method EUA, which is accepted in ICML2026
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What does this PR do?
Fixes # (issue)
We update the hyperparameter settings of SatImp.
Based on feedback from several colleagues, we noticed that the default hyperparameter settings for SatImp were not effective in the first version. This is because these settings were evaluated using the old TOFU code rather than OpenUnlearning. We also found that, under the same hyperparameter settings, the model obtained in OpenUnlearning with 10 epochs is similar to the one obtained in TOFU with 10 epochs but with early stopping at the 2nd epoch.
We add ES metric evaluation on the retain dataset.
As mentioned in the original paper, ES is also used for retain data. Thus, we added this option to the metric module. This makes it convenient to use
ES_forgetandES_retainto compute the forget-retain trade-off with a deviation score.Before submitting