# Hospital Course Summarization on MIMIC-III-ext-VeriFact-BHC 120 prompts (test)

Source: [Hospital Course Summarization on MIMIC-III-ext-VeriFact-BHC 120 prompts (test)](<https://www.sota2.com/research/sota/hospital-course-summarization-on-mimic-iii-ext-verifact-bhc-120-prompts-test>)

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

Data updated: 2026-03-12T08:44:49.913Z

Evaluation of Hospital Course Summarization performance on the specific MIMIC-III-ext-VeriFact-BHC 120 prompts test set, assessed using LLM-based verification metrics (e.g., NS-rate, Supportiveness) and text length.

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 | Hospital Course Summarization |
| Dataset (including any stated split) | MIMIC-III-ext-VeriFact-BHC 120 prompts (test) |
| Metric | Local Verifier NS-rate |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Local Verifier NS-rate](<https://www.sota2.com/research/sota/hospital-course-summarization-on-mimic-iii-ext-verifact-bhc-120-prompts-test.md?metric=Local%20Verifier%20NS-rate>): Higher is better; unit: PERCENT.

- [Local Verifier #NS Count](<https://www.sota2.com/research/sota/hospital-course-summarization-on-mimic-iii-ext-verifact-bhc-120-prompts-test.md?metric=Local%20Verifier%20%23NS%20Count>): Lower is better; unit: SCALAR.

- [Local Verifier #Supp Count](<https://www.sota2.com/research/sota/hospital-course-summarization-on-mimic-iii-ext-verifact-bhc-120-prompts-test.md?metric=Local%20Verifier%20%23Supp%20Count>): Higher is better; unit: SCALAR.

- [GPT-4o NS-rate](<https://www.sota2.com/research/sota/hospital-course-summarization-on-mimic-iii-ext-verifact-bhc-120-prompts-test.md?metric=GPT-4o%20NS-rate>): Higher is better; unit: PERCENT.

- [GPT-4o #NS Count](<https://www.sota2.com/research/sota/hospital-course-summarization-on-mimic-iii-ext-verifact-bhc-120-prompts-test.md?metric=GPT-4o%20%23NS%20Count>): Lower is better; unit: SCALAR.

- [GPT-4o #Supp Count](<https://www.sota2.com/research/sota/hospital-course-summarization-on-mimic-iii-ext-verifact-bhc-120-prompts-test.md?metric=GPT-4o%20%23Supp%20Count>): Higher is better; unit: SCALAR.

- [Validity Rate](<https://www.sota2.com/research/sota/hospital-course-summarization-on-mimic-iii-ext-verifact-bhc-120-prompts-test.md?metric=Validity%20Rate>): Higher is better; unit: PERCENT.

- [Summary Length (Chars)](<https://www.sota2.com/research/sota/hospital-course-summarization-on-mimic-iii-ext-verifact-bhc-120-prompts-test.md?metric=Summary%20Length%20(Chars)>): Higher is better; unit: SIZE.

## Recorded results

| Method | Local Verifier NS-rate | Result date | Report paper |
| --- | --- | --- | --- |
| Base (Llama-3.1) | 10.7 | Not recorded | 69b248a9d030084c22e2f9f6 |
| SFT Baseline | 10.1 | Not recorded | 69b248a9d030084c22e2f9f6 |
| Verifier Rerank | 3.4 | Not recorded | 69b248a9d030084c22e2f9f6 |
| DPO (VERI-DPO) | 1.9 | Not recorded | 69b248a9d030084c22e2f9f6 |

### Base (Llama-3.1)

Local Verifier NS-rate: 10.7. Result ID: 69b27ce86dbb7075b95d2bcb.

[Reporting paper](<https://www.sota2.com/research/paper/69b248a9d030084c22e2f9f6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Backbone":"Llama-3.1"}. This record does not assert independent reproduction.

### SFT Baseline

Local Verifier NS-rate: 10.1. Result ID: 69b27ce86dbb7075b95d2bcc.

[Reporting paper](<https://www.sota2.com/research/paper/69b248a9d030084c22e2f9f6.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

### Verifier Rerank

Local Verifier NS-rate: 3.4. Result ID: 69b27ce86dbb7075b95d2bcd.

[Reporting paper](<https://www.sota2.com/research/paper/69b248a9d030084c22e2f9f6.md>)

Recorded evaluation configuration: {"Reranking Factor (K)":"8"}. This record does not assert independent reproduction.

### DPO (VERI-DPO)

Local Verifier NS-rate: 1.9. Result ID: 69b27ce86dbb7075b95d2bce.

[Reporting paper](<https://www.sota2.com/research/paper/69b248a9d030084c22e2f9f6.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69b248a9d030084c22e2f9f6.md>)

Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction.

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