Research map / Large language models / Retrieval and knowledge
AgriEval: A Comprehensive Chinese Agricultural Benchmark for Large Language Models
Lian Yan, Haotian Wang, Chen Tang, Haifeng Liu, Tianyang Sun, Liangliang Liu, Yi Guan, Jingchi Jiang
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 |
|---|---|---|---|
| add chain-of-thought prompting (ablation) | Instruct LLMs · AgriEval multiple-choice overall | accuracy: average -3.51% | Hurt |
| add chain-of-thought prompting (ablation) | Instruct LLMs · AgriEval numerical reasoning | accuracy: +9.81% average; Qwen2.5-14B 49.21 -> 64.99 | Improved |
| add 5-shot examples (ablation) | Instruct LLMs · AgriEval multiple-choice overall | accuracy: GLM-4-9B 43.72 -> 40.07; Qwen2.5-14B 49.53 -> 52.07 | Mixed |
| add RAG with Chinese Wikipedia (ablation) | LLMs · AgriEval 1,000-sample subset | accuracy: about +4.0% average | Improved |
| instruction tuning vs base models (ablation) | LLMs · AgriEval | accuracy: average +10.60% | Improved |
| scale Qwen2.5 from 3B to 72B | Qwen2.5-Instruct · AgriEval | accuracy: 39.67 -> 60.32 | Improved |
| place correct answers in later option positions (ablation) | LLMs · AgriEval shuffled options | accuracy: 58.50% later vs 37.84% selected | Hurt |
Closest papers by meaning
- Empowering and Assessing the Utility of Large Language Models in Crop Science (NeurIPS 2024 · similarity 0.89)
- AgMMU: A Comprehensive Agricultural Multimodal Understanding Benchmark (NeurIPS 2025 · similarity 0.88)
- AgroOmni: A Large-Scale Multi-view Agricultural Dataset for Cross-Scale Multimodal Reasoning (NeurIPS 2026 · similarity 0.87)
- SuperGPQA: Scaling LLM Evaluation across 285 Graduate Disciplines (NeurIPS 2025 · similarity 0.86)
- CK12: A Rounded K12 Knowledge Graph Based Benchmark for Chinese Holistic Cognition Evaluation (AAAI 2024 · similarity 0.85)
- MME-SCI: A Comprehensive and Challenging Science Benchmark for Multimodal Large Language Models (AAAI 2026 · similarity 0.85)
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.
