Research map / Large language models / Retrieval and knowledge
Benchmarking Large Language Models on CMExam - A comprehensive Chinese Medical Exam Dataset
Junling Liu, Peilin Zhou, Yining Hua, Dading Chong, Zhongyu Tian, Andrew Liu, Helin Wang, Chenyu You, Zhenhua Guo, Zhu Lei, Michael Lingzhi Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| fine-tune on CMExam training set (P-tuning v2) | ChatGLM-6B · CMExam prediction | accuracy: 26.3 -> 45.3 | Improved |
| fine-tune on CMExam training set (LoRA) | LLaMA-7B · CMExam prediction | accuracy: 0.4 -> 18.3 | Improved |
| fine-tune on CMExam (LoRA) | Huatuo-7B · CMExam prediction | accuracy: 12.9 -> 28.6 | Improved |
| fine-tune on CMExam (LoRA) | Vicuna-7B · CMExam prediction | accuracy: 5.0 -> 27.3 | Improved |
| fine-tune on CMExam (LoRA) | Alpaca-7B · CMExam prediction | accuracy: 8.5 -> 21.1 | Improved |
| fine-tune on CMExam | ChatGLM-6B · CMExam explanation | BLEU-4: 5.00 -> 18.94 | Improved |
| add few-shot prompt (ablation) | GPT-4 · CMExam prediction | accuracy: 61.6 -> 62.0 | No clear effect |
| add chain-of-thought prompt (ablation) | GPT-4 · CMExam prediction | accuracy: 61.6 -> 61.6 | No clear effect |
| add few-shot prompt (ablation) | GPT-3.5 · CMExam prediction | accuracy: 46.4 -> 45.3 | No clear effect |
| add chain-of-thought prompt (ablation) | GPT-3.5 · CMExam prediction | accuracy: 46.4 -> 47.9 | Improved |
| add few-shot prompt (ablation) | GPT-4 · CMExam explanation | BLEU-1: 0.17 -> 5.95 | Improved |
| add chain-of-thought prompt (ablation) | GPT-4 · CMExam explanation | BLEU-1: 0.17 -> 7.29 | Improved |
Closest papers by meaning
- MedBench: A Large-Scale Chinese Benchmark for Evaluating Medical Large Language Models (AAAI 2024 · similarity 0.93)
- Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks (NeurIPS 2026 · similarity 0.91)
- MedJourney: Benchmark and Evaluation of Large Language Models over Patient Clinical Journey (NeurIPS 2024 · similarity 0.91)
- TCM-Ladder: A Benchmark for Multimodal Question Answering on Traditional Chinese Medicine (NeurIPS 2025 · similarity 0.91)
- Zhongjing: Enhancing the Chinese Medical Capabilities of Large Language Model through Expert Feedback and Real-World Multi-Turn Dialogue (AAAI 2024 · similarity 0.89)
- MedCalc-Bench: Evaluating Large Language Models for Medical Calculations (NeurIPS 2024 · similarity 0.89)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
