Research map / Large language models / Retrieval and knowledge
CliCARE: Grounding Large Language Models in Clinical Guidelines for Decision Support over Longitudinal Cancer Electronic Health Records
Dongchen Li, Jitao Liang, Wei Li, Xiaoyu Wang, Longbing Cao, Kun Yu
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 |
|---|---|---|---|
| apply CliCARE (TKG plus guideline alignment) instead of standard RAG | Qwen3-8B · CancerEHR clinical summary | LLM-judge score: 1.485 -> 3.173 (+1.688) | Improved |
| apply CliCARE instead of standard RAG | Gemini 2.5 Pro · CancerEHR clinical summary | LLM-judge score: 2.735 -> 4.976 (+2.241) | Improved |
| apply CliCARE instead of standard RAG | GPT-4.1 · MIMIC-Cancer clinical summary | LLM-judge score: 4.419 -> 4.737 (+0.318) | Improved |
| apply CliCARE instead of standard RAG | Mistral-Instruct-v0.1-7B · MIMIC-Cancer clinical recommendation | LLM-judge score: 2.115 -> 2.071 (-0.044) | Hurt |
| add alignment expansion (ablation) | CliCARE (Qwen3-8B) · MIMIC-Cancer clinical summary | LLM-judge score: 2.075 -> 2.575 | Improved |
| add LLM-based reranking (ablation) | CliCARE (Qwen3-8B) · CancerEHR clinical summary | LLM-judge score: 2.857 -> 3.012 | Improved |
| add LLM-based reranking (ablation) | CliCARE (Gemini 2.5 Pro) · MIMIC-Cancer clinical summary | LLM-judge score: 3.774 -> 3.737 | Hurt |
| add TKG-based compression (ablation) | CliCARE (Qwen3-8B) · CancerEHR clinical summary | LLM-judge score: 1.485 -> 2.857 | Improved |
| add TKG-based compression (ablation) | CliCARE (Qwen3-8B) · MIMIC-Cancer clinical summary | LLM-judge score: 2.475 -> 2.000 | Hurt |
| add TKG-based compression (ablation) | CliCARE (Gemini 2.5 Pro) · CancerEHR clinical summary | LLM-judge score: 2.735 -> 4.542 | Improved |
| longer EHR input (long vs short records) | CliCARE (Qwen3-8B) · CancerEHR clinical summary | LLM-judge score: 3.228 -> 2.976 | Hurt |
| longer EHR input (long vs short records) | CliCARE (Gemini 2.5 Pro) · CancerEHR clinical summary | LLM-judge score: 4.982 -> 4.988 | No clear effect |
Closest papers by meaning
- Reasoning-Enhanced Healthcare Predictions with Knowledge Graph Community Retrieval (ICLR 2025 · similarity 0.87)
- MedJourney: Benchmark and Evaluation of Large Language Models over Patient Clinical Journey (NeurIPS 2024 · similarity 0.86)
- Context Clues: Evaluating Long Context Models for Clinical Prediction Tasks on EHR Data (ICLR 2025 · similarity 0.86)
- EHRNoteQA: An LLM Benchmark for Real-World Clinical Practice Using Discharge Summaries (NeurIPS 2024 · similarity 0.86)
- SM3-Text-to-Query: Synthetic Multi-Model Medical Text-to-Query Benchmark (NeurIPS 2024 · similarity 0.86)
- HC-LLM: Historical-Constrained Large Language Models for Radiology Report Generation (AAAI 2025 · similarity 0.86)
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.
