Note-table consistency verification on EHRCon Physician notes
79.58RecallEHR-INSPECTOR
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| EHR-INSPECTORBase LLM=Gemini 2.5 Flash, Evaluation Setting=Lenient2026.05 | 79.58 | — | — | — | — | 71.81 | |
| EHR-INSPECTORBase LLM=Gemini 2.5 Flash, Evaluation Setting=Harsh2026.05 | 73.14 | — | — | — | — | 66.11 | |
| EHR-INSPECTORBase LLM=Qwen3 32B, Evaluation Setting=Lenient2026.05 | 55.34 | — | — | — | — | 65.21 | |
| EHR-INSPECTORBase LLM=Qwen3 32B, Evaluation Setting=Harsh2026.05 | 49.21 | — | — | — | — | 58.98 | |
| EHR-INSPECTORBase LLM=MedGemma 27B, Evaluation Setting=Lenient2026.05 | 28.63 | — | — | — | — | 33.44 | |
| CheckEHRBase LLM=Gemini 2.5 Flash, Evaluation Setting=Lenient2026.05 | 28.05 | — | — | — | — | 35.86 | |
| CheckEHRBase LLM=Qwen3 32B, Evaluation Setting=Lenient2026.05 | 27.55 | — | — | — | — | 48.64 | |
| EHR-INSPECTORBase LLM=GPT-OSS 20B, Evaluation Setting=Lenient2026.05 | 27.53 | — | — | — | — | 55.01 | |
| EHR-INSPECTORBase LLM=MedGemma 27B, Evaluation Setting=Harsh2026.05 | 26.93 | — | — | — | — | 30.55 | |
| CheckEHRBase LLM=Gemini 2.5 Flash, Evaluation Setting=Harsh2026.05 | 26.38 | — | — | — | — | 30.61 | |
| CheckEHRBase LLM=Qwen3 32B, Evaluation Setting=Harsh2026.05 | 25.31 | — | — | — | — | 43.95 | |
| CheckEHRBase LLM=GPT-OSS 20B, Evaluation Setting=Lenient2026.05 | 24.91 | — | — | — | — | 39.23 | |
| CheckEHRBase LLM=GPT-OSS 20B, Evaluation Setting=Harsh2026.05 | 23 | — | — | — | — | 35.3 | |
| EHR-INSPECTORBase LLM=GPT-OSS 20B, Evaluation Setting=Harsh2026.05 | 22.17 | — | — | — | — | 49.33 | |
| CheckEHRBase LLM=MedGemma 27B, Evaluation Setting=Lenient2026.05 | 6.6 | — | — | — | — | 35.09 | |
| CheckEHRBase LLM=MedGemma 27B, Evaluation Setting=Harsh2026.05 | 6.1 | — | — | — | — | 32.71 | |
| CheckEHR2026.05 | — | 60.65 | 47.29 | 59.22 | 43.94 | — | |
| EHR-INSPECTOR2026.05 | — | 86.02 | 93.09 | 79.66 | 84.93 | — |