MiCEval
An automatic evaluation framework for Multimodal Chain-of-Thought.
Install / Use
/learn @alenai97/MiCEvalREADME
MiCEval: Unveiling Multimodal Chain of Thought's Quality via Image Description and Reasoning
The MiCEval dataset is stored in dataset/MiCEval_hard.jsonl and dataset/MiCEval_Normal.jsonl.
The images of the MiCEval dataset are stored in the dataset/image folder.
To reproduce the results of the verifier experiment, please run scripts/verifier_zeroshot.sh and scripts/verifier_fewshot.sh.
The code for the evaluator experiment will be updated soon.
Related Skills
node-connect
349.9kDiagnose OpenClaw node connection and pairing failures for Android, iOS, and macOS companion apps
frontend-design
109.8kCreate distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, or applications. Generates creative, polished code that avoids generic AI aesthetics.
openai-whisper-api
349.9kTranscribe audio via OpenAI Audio Transcriptions API (Whisper).
qqbot-media
349.9kQQBot 富媒体收发能力。使用 <qqmedia> 标签，系统根据文件扩展名自动识别类型（图片/语音/视频/文件）。
