Research map / Large language models / Retrieval and knowledge
EHRSQL: A Practical Text-to-SQL Benchmark for Electronic Health Records
Gyubok Lee, Hyeonji Hwang, Seongsu Bae, Yeonsu Kwon, Woncheol Shin, Seongjun Yang, Minjoon Seo, Jong-Yeup Kim, Edward Choi
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 entropy-based refusal with percentile-based threshold (ablation) | T5-base · EHRSQL MIMIC-III test | F1exe: 76.6 -> 89.9 | Improved |
| add entropy-based refusal with clustering-based threshold (ablation) | T5-base · EHRSQL MIMIC-III test | F1exe: 76.6 -> 88.1 | Improved |
| add entropy-based refusal with percentile-based threshold (ablation) | T5-base · EHRSQL eICU test | F1exe: 76.5 -> 89.5 | Improved |
| percentile-based instead of clustering-based threshold (ablation) | T5-base · EHRSQL MIMIC-III test | F1exe: 88.1 -> 89.9 | Improved |
| add schema serialization to input (ablation) | T5-base · EHRSQL MIMIC-III test | F1exe: 89.9 -> 89.1 | No clear effect |
| add schema serialization to input (ablation) | T5-base · EHRSQL eICU test | F1exe: 89.5 -> 88.9 | No clear effect |
| zero-shot transfer of Spider-trained GAP to EHRSQL instead of MIMICSQL | GAP · healthcare text-to-SQL | execution accuracy (%): 16.4 -> 4.7 | Hurt |
Closest papers by meaning
- SM3-Text-to-Query: Synthetic Multi-Model Medical Text-to-Query Benchmark (NeurIPS 2024 · similarity 0.90)
- EHRNoteQA: An LLM Benchmark for Real-World Clinical Practice Using Discharge Summaries (NeurIPS 2024 · similarity 0.87)
- From Conversation to Query Execution: Benchmarking User and Tool Interactions for EHR Database Agents (ICLR 2026 · similarity 0.87)
- MedAlign: A Clinician-Generated Dataset for Instruction Following with Electronic Medical Records (AAAI 2024 · similarity 0.86)
- EHRCon: Dataset for Checking Consistency between Unstructured Notes and Structured Tables in Electronic Health Records (NeurIPS 2024 · similarity 0.86)
- Context Clues: Evaluating Long Context Models for Clinical Prediction Tasks on EHR Data (ICLR 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.
