# Note-table consistency verification on EHRCon Discharge notes

Source: [Note-table consistency verification on EHRCon Discharge notes](<https://www.sota2.com/research/sota/note-table-consistency-verification-on-ehrcon-discharge-notes>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-27T08:20:33.544Z

Evaluation of model performance on verifying consistency between notes and associated tables within EHRCon Discharge notes, measured using Lenient and Harsh Precision/Recall scores.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Note-table consistency verification |
| Dataset (including any stated split) | EHRCon Discharge notes |
| Metric | Recall |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 18 of 18 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Recall](<https://www.sota2.com/research/sota/note-table-consistency-verification-on-ehrcon-discharge-notes.md?metric=Recall>): Higher is better; unit: PERCENT.

- [Lenient Recall](<https://www.sota2.com/research/sota/note-table-consistency-verification-on-ehrcon-discharge-notes.md?metric=Lenient%20Recall>): Higher is better; unit: PERCENT.

- [Lenient Precision](<https://www.sota2.com/research/sota/note-table-consistency-verification-on-ehrcon-discharge-notes.md?metric=Lenient%20Precision>): Higher is better; unit: PERCENT.

- [Harsh Recall](<https://www.sota2.com/research/sota/note-table-consistency-verification-on-ehrcon-discharge-notes.md?metric=Harsh%20Recall>): Higher is better; unit: PERCENT.

- [Harsh Precision](<https://www.sota2.com/research/sota/note-table-consistency-verification-on-ehrcon-discharge-notes.md?metric=Harsh%20Precision>): Higher is better; unit: PERCENT.

- [Precision](<https://www.sota2.com/research/sota/note-table-consistency-verification-on-ehrcon-discharge-notes.md?metric=Precision>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Recall | Result date | Report paper |
| --- | --- | --- | --- |
| EHR-INSPECTOR | 77.36 | Not recorded | 6a167aaddbd7150506f800db |
| EHR-INSPECTOR | 69.28 | Not recorded | 6a167aaddbd7150506f800db |
| EHR-INSPECTOR | 56.18 | Not recorded | 6a167aaddbd7150506f800db |
| EHR-INSPECTOR | 50.89 | Not recorded | 6a167aaddbd7150506f800db |
| CheckEHR | 48.8 | Not recorded | 6a167aaddbd7150506f800db |
| EHR-INSPECTOR | 48.68 | Not recorded | 6a167aaddbd7150506f800db |
| CheckEHR | 44.89 | Not recorded | 6a167aaddbd7150506f800db |
| EHR-INSPECTOR | 42.74 | Not recorded | 6a167aaddbd7150506f800db |
| EHR-INSPECTOR | 30 | Not recorded | 6a167aaddbd7150506f800db |
| CheckEHR | 29.4 | Not recorded | 6a167aaddbd7150506f800db |
| EHR-INSPECTOR | 26.99 | Not recorded | 6a167aaddbd7150506f800db |
| CheckEHR | 26.5 | Not recorded | 6a167aaddbd7150506f800db |
| CheckEHR | 26.05 | Not recorded | 6a167aaddbd7150506f800db |
| CheckEHR | 23.04 | Not recorded | 6a167aaddbd7150506f800db |
| CheckEHR | 17.95 | Not recorded | 6a167aaddbd7150506f800db |
| CheckEHR | 16.41 | Not recorded | 6a167aaddbd7150506f800db |

### EHR-INSPECTOR

Recall: 77.36. Result ID: 6a16a7ebfbc373592c0fb5ba.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

Recorded evaluation configuration: {"Base LLM":"Gemini 2.5 Flash","Evaluation Setting":"Lenient"}. This record does not assert independent reproduction.

### EHR-INSPECTOR

Recall: 69.28. Result ID: 6a16a7ebfbc373592c0fb5c2.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

Recorded evaluation configuration: {"Base LLM":"Gemini 2.5 Flash","Evaluation Setting":"Harsh"}. This record does not assert independent reproduction.

### EHR-INSPECTOR

Recall: 56.18. Result ID: 6a16a7ebfbc373592c0fb5b8.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

Recorded evaluation configuration: {"Base LLM":"Qwen3 32B","Evaluation Setting":"Lenient"}. This record does not assert independent reproduction.

### EHR-INSPECTOR

Recall: 50.89. Result ID: 6a16a7ebfbc373592c0fb5c0.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

Recorded evaluation configuration: {"Base LLM":"Qwen3 32B","Evaluation Setting":"Harsh"}. This record does not assert independent reproduction.

### CheckEHR

Recall: 48.8. Result ID: 6a16a7ebfbc373592c0fb5b9.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76731e209131cea6d110.md>)

Recorded evaluation configuration: {"Base LLM":"Gemini 2.5 Flash","Evaluation Setting":"Lenient"}. This record does not assert independent reproduction.

### EHR-INSPECTOR

Recall: 48.68. Result ID: 6a16a7ebfbc373592c0fb5b6.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

Recorded evaluation configuration: {"Base LLM":"MedGemma 27B","Evaluation Setting":"Lenient"}. This record does not assert independent reproduction.

### CheckEHR

Recall: 44.89. Result ID: 6a16a7ebfbc373592c0fb5c1.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76731e209131cea6d110.md>)

Recorded evaluation configuration: {"Base LLM":"Gemini 2.5 Flash","Evaluation Setting":"Harsh"}. This record does not assert independent reproduction.

### EHR-INSPECTOR

Recall: 42.74. Result ID: 6a16a7ebfbc373592c0fb5be.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

Recorded evaluation configuration: {"Base LLM":"MedGemma 27B","Evaluation Setting":"Harsh"}. This record does not assert independent reproduction.

### EHR-INSPECTOR

Recall: 30. Result ID: 6a16a7ebfbc373592c0fb5b4.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

Recorded evaluation configuration: {"Base LLM":"GPT-OSS 20B","Evaluation Setting":"Lenient"}. This record does not assert independent reproduction.

### CheckEHR

Recall: 29.4. Result ID: 6a16a7ebfbc373592c0fb5b3.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76731e209131cea6d110.md>)

Recorded evaluation configuration: {"Base LLM":"GPT-OSS 20B","Evaluation Setting":"Lenient"}. This record does not assert independent reproduction.

### EHR-INSPECTOR

Recall: 26.99. Result ID: 6a16a7ebfbc373592c0fb5bc.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

Recorded evaluation configuration: {"Base LLM":"GPT-OSS 20B","Evaluation Setting":"Harsh"}. This record does not assert independent reproduction.

### CheckEHR

Recall: 26.5. Result ID: 6a16a7ebfbc373592c0fb5bb.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76731e209131cea6d110.md>)

Recorded evaluation configuration: {"Base LLM":"GPT-OSS 20B","Evaluation Setting":"Harsh"}. This record does not assert independent reproduction.

### CheckEHR

Recall: 26.05. Result ID: 6a16a7ebfbc373592c0fb5b7.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76731e209131cea6d110.md>)

Recorded evaluation configuration: {"Base LLM":"Qwen3 32B","Evaluation Setting":"Lenient"}. This record does not assert independent reproduction.

### CheckEHR

Recall: 23.04. Result ID: 6a16a7ebfbc373592c0fb5bf.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76731e209131cea6d110.md>)

Recorded evaluation configuration: {"Base LLM":"Qwen3 32B","Evaluation Setting":"Harsh"}. This record does not assert independent reproduction.

### CheckEHR

Recall: 17.95. Result ID: 6a16a7ebfbc373592c0fb5b5.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76731e209131cea6d110.md>)

Recorded evaluation configuration: {"Base LLM":"MedGemma 27B","Evaluation Setting":"Lenient"}. This record does not assert independent reproduction.

### CheckEHR

Recall: 16.41. Result ID: 6a16a7ebfbc373592c0fb5bd.

[Reporting paper](<https://www.sota2.com/research/paper/6a167aaddbd7150506f800db.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa76731e209131cea6d110.md>)

Recorded evaluation configuration: {"Base LLM":"MedGemma 27B","Evaluation Setting":"Harsh"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.
