Skip to content

Fix LLM4Mat eval (SciReasoner)#2366

Merged
Myhs-phz merged 2 commits intoopen-compass:mainfrom
yusun-nlp:scireasoner_v1
Dec 23, 2025
Merged

Fix LLM4Mat eval (SciReasoner)#2366
Myhs-phz merged 2 commits intoopen-compass:mainfrom
yusun-nlp:scireasoner_v1

Conversation

@yusun-nlp
Copy link
Contributor

Motivation

Fix the post-processing rules for the LLM4Mat subtask, allowing it to match more valid answers.

Modification

When the output is True/False, answers with similar expressions are allowed; when the output is in dict form, both single and double quotes are allowed.

@Myhs-phz Myhs-phz merged commit 26ae808 into open-compass:main Dec 23, 2025
7 checks passed
iamkaia pushed a commit to iamkaia/opencompass that referenced this pull request Feb 4, 2026
* fix LLM4Mat regex

* fix LLM4Mat regex
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants