Research map / Large language models / AI and society
NurValues: Real-World Nursing Values Evaluation for Large Language Models in Clinical Context
Ben Yao, Qiuchi Li, Yazhou Zhang, Yang siyu, Bohan Zhang, Prayag Tiwari, Jing Qin
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) | DeepSeek-V3 · NurValues Hard-Level | Ma-F1: 34.29 -> 57.32 | Improved |
| add chain-of-thought prompting (ablation) | DeepSeek-V3 · NurValues Easy-Level | Ma-F1: 94.55 -> 93.64 | Hurt |
| add chain-of-thought prompting (ablation) | HuatuoGPT-o1-72B · NurValues Easy-Level | Ma-F1: 89.88 -> 62.14 | Hurt |
| add chain-of-thought prompting (ablation) | Qwen2.5-72B-Instruct · NurValues Easy-Level | Ma-F1: 93.31 -> 94.68 | Improved |
| add self-consistency (ablation) | Llama-3-8B-Instruct · NurValues Hard-Level | Ma-F1: 6.66 -> 28.05 | Improved |
| add 10-shot examples (ablation) | Llama-3.1-70B-Instruct · NurValues Hard-Level | Ma-F1: 24.96 -> 32.99 | Improved |
| add 6-shot examples (ablation) | Llama-3.1-70B-Instruct · NurValues Easy-Level | Ma-F1: 73.54 -> 92.14 | Improved |
| use general instead of medical-tuned LLMs | LLMs on NurValues · NurValues Hard-Level | Ma-F1: 18.42 -> 37.84 | Improved |
| use reasoning models over non-reasoning | Gemini-2.5-Pro, o1, DeepSeek-R1, QwQ-Plus · NurValues Hard-Level | accuracy | No clear effect |
| regenerate Hard-Level with a different LLM (DeepSeek-V3) (ablation) | Claude 3.7 Sonnet · NurValues Hard-Level | Ma-F1: 79.89 -> 68.35 | No clear effect |
| increase shots 2 to 10 (ablation) | DeepSeek-V3 · NurValues Hard-Level | Ma-F1: 34.44 -> 42.95 | Improved |
Closest papers by meaning
- What Does the AI Doctor Value? Auditing Pluralism in the Clinical Ethics of Language Models (NeurIPS 2026 · similarity 0.86)
- Position: Beyond Assistance – Reimagining LLMs as Ethical and Adaptive Co-Creators in Mental Health Care (ICML 2025 · similarity 0.84)
- Value Imprint: A Technique for Auditing the Human Values Embedded in RLHF Datasets (NeurIPS 2024 · similarity 0.84)
- CLAVE: An Adaptive Framework for Evaluating Values of LLM Generated Responses (NeurIPS 2024 · similarity 0.84)
- MedAlign: A Clinician-Generated Dataset for Instruction Following with Electronic Medical Records (AAAI 2024 · similarity 0.83)
- MedJourney: Benchmark and Evaluation of Large Language Models over Patient Clinical Journey (NeurIPS 2024 · similarity 0.83)
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.
