Conversation
|
Ashish seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account. You have signed the CLA already but the status is still pending? Let us recheck it. |
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This PR implements the BioASQ environment for large-scale biomedical semantic question answering, evaluating models' ability to generate comprehensive, expert-level answers grounded in scientific literature.
Dataset Integration: Uses the kroshan/BioASQ dataset from HuggingFace containing 5,729 biomedical questions with ideal answers and supporting evidence, following BioASQ 2025 Task 13b specifications.
Evaluation Framework: Implements LLM-as-judge evaluation with 4-dimensional scoring exactly as specified in official BioASQ manual assessment criteria:
Precision (1-5): Accuracy and relevance of biomedical facts
Recall (1-5): Coverage of important biomedical concepts
Repetition (1-5): Redundancy assessment (inverse scoring)
Readability (1-5): Clarity and professional organization
Answer Formats: Supports both XML (...) and BOXED answer formatting to match med-lm-envs conventions.
Judge Integration: Uses JudgeRubric with configurable judge models (default: gpt-4o-mini) and follows the exact async judge pattern from other med-lm-envs environments like aci_bench.
Technical Implementation: Follows all med-lm-envs patterns including proper data parsing, normalized scoring (0-1 range), CLI integration with environment-specific arguments, and comprehensive documentation with dataset examples and statistics.