AI Paper Atlas

Research map / Large language models / Agents, code and math

C-Eval: A Multi-Level Multi-Discipline Chinese Evaluation Suite for Foundation Models

Yuzhen Huang, Yuzhuo Bai, Zhihao Zhu, Junlei Zhang, Jinghan Zhang, Tangjun Su, Junteng Liu, Chuancheng Lv, Yikai Zhang, jiayi lei, Yao Fu, Maosong Sun and 1 more

NeurIPS 2023 · Large language models › Agents, code and math · approach: Mathematical and structural reasoning benchmarks · 92 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use five-shot answer-only prompting instead of zero-shot answer-only prompting (ablation)GPT-4 · C-Eval testaverage accuracy (%): 66.4 -> 68.7Improved
Use five-shot answer-only prompting instead of zero-shot answer-only prompting (ablation)ChatGPT · C-Eval testaverage accuracy (%): 51.0 -> 54.4Improved
Use five-shot answer-only prompting instead of zero-shot answer-only prompting (ablation)Claude-v1.3 · C-Eval testaverage accuracy (%): 50.5 -> 54.2Improved
Use five-shot answer-only prompting instead of zero-shot answer-only prompting (ablation)GLM-130B · C-Eval testaverage accuracy (%): 44.0 -> 40.3Hurt
Use five-shot answer-only prompting instead of zero-shot answer-only prompting (ablation)Bloomz-mt · C-Eval testaverage accuracy (%): 44.3 -> 39.0Hurt
Use five-shot answer-only prompting instead of zero-shot answer-only prompting (ablation)ChatGLM-6B · C-Eval testaverage accuracy (%): 38.9 -> 34.5Hurt
Use five-shot answer-only prompting instead of zero-shot answer-only prompting (ablation)MOSS · C-Eval testaverage accuracy (%): 33.1 -> 31.1Hurt
Use five-shot answer-only prompting instead of zero-shot answer-only prompting (ablation)Chinese-Alpaca-13B · C-Eval testaverage accuracy (%): 30.9 -> 26.7Hurt
Add five-shot chain-of-thought prompting instead of five-shot answer-only prompting (ablation)GPT-4 · C-Eval testaverage accuracy (%): 68.7 -> 68.3Hurt
Add five-shot chain-of-thought prompting instead of five-shot answer-only prompting (ablation)ChatGPT · C-Eval testaverage accuracy (%): 54.4 -> 50.0Hurt
Add five-shot chain-of-thought prompting instead of five-shot answer-only prompting (ablation)GLM-130B · C-Eval testaverage accuracy (%): 40.3 -> 28.8Hurt
Add five-shot chain-of-thought prompting instead of five-shot answer-only prompting (ablation)GPT-4 · C-Eval Hardaverage accuracy (%): 54.9 -> 56.8Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy