[Benchmark] Add support for CAPEval - #1631
Closed
liuzhipenggg wants to merge 1 commit into
Closed
Conversation
liuzhipenggg
force-pushed
the
add-capeval
branch
2 times, most recently
from
August 14, 2026 14:55
1fffac0 to
3b634d3
Compare
Author
|
Superseded by #1633 (rebased onto current |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
Add CAPEval as an image caption benchmark.
The dataset has 300 images. The model produces a caption; a judge LLM scores it against checklist items and reports Coverage (C) and Precision (P), with Scene & Object (SO) / People & Activity (PA) / Text & Interface (TI) / Design & Knowledge (DK) splits.
Changes
CAPEvalinIMAGE_DATASETvlmeval/dataset/capeval.pyvlmeval/dataset/utils/capeval/$LMUData/CAPEval/Implemented as
ImageBaseDataset+build_judge. No changes torun.py, model backends, or shared judge code.Usage
--judge defaults to qwen-72b if omitted.
References
https://github.com/liuzhipenggg/CAPEval
https://huggingface.co/datasets/LiuzhipengUCAS/CAPEval