Add hallucination multicalibrator, with example benchmark #152
Conversation
This file contains 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
wistuba
reviewed
Nov 16, 2023
benchmarks/hallucination/mmlu/run.py
Outdated
| for task in task_list | ||
| ] | ||
|
|
||
| answer_map = {a: i for i, a in enumerate(auc)} |
There was a problem hiding this comment.
choices seem to be in list("ABCD"). why has the answer map more options?
There was a problem hiding this comment.
Tries to be more generic, but you're right, can be restricted to "ABCD".
fortuna/hallucination/base.py
Outdated
| } | ||
|
|
||
| with torch.no_grad(): | ||
| __logits = self.generative_model( |
|
|
||
|
|
||
| class TestScoringModel(unittest.TestCase): | ||
| def test_score(self): |
There was a problem hiding this comment.
can't you compare against exact values?
| import re | ||
|
|
||
|
|
||
| def string_cleaner(text: str) -> str: |
There was a problem hiding this comment.
great candidate for a function to have unit tests
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
Add the
HallucinationMulticalibrator. This is able to take in input a generative model, a tokenizer and some data, then fit the calibrator and predict calibrated probabilities of hallucinations.