# Rare-disease diagnosis on CROSSRARE-BENCH Overall 1.0 (test)

Source: [Rare-disease diagnosis on CROSSRARE-BENCH Overall 1.0 (test)](<https://www.sota2.com/research/sota/rare-disease-diagnosis-on-crossrare-bench-overall-1-0-test>)

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

Data updated: 2026-06-15T07:02:13.272Z

Evaluation of rare-disease diagnosis performance using the test split of the CROSSRARE-BENCH Overall 1.0 dataset, measured by Accuracy, F1, and Judge 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 | Rare-disease diagnosis |
| Dataset (including any stated split) | CROSSRARE-BENCH Overall 1.0 (test) |
| Metric | Accuracy |
| 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

- [Accuracy](<https://www.sota2.com/research/sota/rare-disease-diagnosis-on-crossrare-bench-overall-1-0-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/rare-disease-diagnosis-on-crossrare-bench-overall-1-0-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Judge Score](<https://www.sota2.com/research/sota/rare-disease-diagnosis-on-crossrare-bench-overall-1-0-test.md?metric=Judge%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| MedLatentDx | 73 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Struct. RAG | 64 | Not recorded | 6a2f871cd33dfb23e783c010 |
| MedLatentDx | 64 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Global RAG | 62 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Dist. RAG | 61 | Not recorded | 6a2f871cd33dfb23e783c010 |
| LatentMAS | 60 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Raw KV | 59 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Single RAG | 57 | Not recorded | 6a2f871cd33dfb23e783c010 |
| TextMAS | 54 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Global RAG | 47 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Dist. RAG | 45 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Struct. RAG | 42 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Single RAG | 39 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Raw KV | 35 | Not recorded | 6a2f871cd33dfb23e783c010 |
| LatentMAS | 33 | Not recorded | 6a2f871cd33dfb23e783c010 |
| TextMAS | 32 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Zero-shot | 3 | Not recorded | 6a2f871cd33dfb23e783c010 |
| Zero-shot | 1 | Not recorded | 6a2f871cd33dfb23e783c010 |

### MedLatentDx

Accuracy: 73. Result ID: 6a2fa35ffbc373592c2b6641.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Llama-3.2-3B","Collaboration Category":"Latent communication"}. This record does not assert independent reproduction.

### Struct. RAG

Accuracy: 64. Result ID: 6a2fa35ffbc373592c2b6634.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B","Collaboration Category":"Text / structured sharing"}. This record does not assert independent reproduction.

### MedLatentDx

Accuracy: 64. Result ID: 6a2fa35ffbc373592c2b6638.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B","Collaboration Category":"Latent communication"}. This record does not assert independent reproduction.

### Global RAG

Accuracy: 62. Result ID: 6a2fa35ffbc373592c2b6633.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B","Collaboration Category":"Text / structured sharing"}. This record does not assert independent reproduction.

### Dist. RAG

Accuracy: 61. Result ID: 6a2fa35ffbc373592c2b6632.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B","Collaboration Category":"Text / structured sharing"}. This record does not assert independent reproduction.

### LatentMAS

Accuracy: 60. Result ID: 6a2fa35ffbc373592c2b6637.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

[Method origin paper](<https://www.sota2.com/research/paper/692689c6d6b0dd334293064c.md>)

[Implementation](<https://github.com/Gen-Verse/LatentMAS>)

Recorded evaluation configuration: {"Model":"Qwen3-4B","Collaboration Category":"Latent communication"}. This record does not assert independent reproduction.

### Raw KV

Accuracy: 59. Result ID: 6a2fa35ffbc373592c2b6636.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B","Collaboration Category":"Latent communication"}. This record does not assert independent reproduction.

### Single RAG

Accuracy: 57. Result ID: 6a2fa35ffbc373592c2b6631.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B","Collaboration Category":"Non-collaborative"}. This record does not assert independent reproduction.

### TextMAS

Accuracy: 54. Result ID: 6a2fa35ffbc373592c2b6635.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B","Collaboration Category":"Text / structured sharing"}. This record does not assert independent reproduction.

### Global RAG

Accuracy: 47. Result ID: 6a2fa35ffbc373592c2b663c.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Llama-3.2-3B","Collaboration Category":"Text / structured sharing"}. This record does not assert independent reproduction.

### Dist. RAG

Accuracy: 45. Result ID: 6a2fa35ffbc373592c2b663b.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Llama-3.2-3B","Collaboration Category":"Text / structured sharing"}. This record does not assert independent reproduction.

### Struct. RAG

Accuracy: 42. Result ID: 6a2fa35ffbc373592c2b663d.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Llama-3.2-3B","Collaboration Category":"Text / structured sharing"}. This record does not assert independent reproduction.

### Single RAG

Accuracy: 39. Result ID: 6a2fa35ffbc373592c2b663a.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Llama-3.2-3B","Collaboration Category":"Non-collaborative"}. This record does not assert independent reproduction.

### Raw KV

Accuracy: 35. Result ID: 6a2fa35ffbc373592c2b663f.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Llama-3.2-3B","Collaboration Category":"Latent communication"}. This record does not assert independent reproduction.

### LatentMAS

Accuracy: 33. Result ID: 6a2fa35ffbc373592c2b6640.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

[Method origin paper](<https://www.sota2.com/research/paper/692689c6d6b0dd334293064c.md>)

[Implementation](<https://github.com/Gen-Verse/LatentMAS>)

Recorded evaluation configuration: {"Model":"Llama-3.2-3B","Collaboration Category":"Latent communication"}. This record does not assert independent reproduction.

### TextMAS

Accuracy: 32. Result ID: 6a2fa35ffbc373592c2b663e.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Llama-3.2-3B","Collaboration Category":"Text / structured sharing"}. This record does not assert independent reproduction.

### Zero-shot

Accuracy: 3. Result ID: 6a2fa35ffbc373592c2b6630.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Qwen3-4B","Collaboration Category":"Non-collaborative"}. This record does not assert independent reproduction.

### Zero-shot

Accuracy: 1. Result ID: 6a2fa35ffbc373592c2b6639.

[Reporting paper](<https://www.sota2.com/research/paper/6a2f871cd33dfb23e783c010.md>)

Recorded evaluation configuration: {"Model":"Llama-3.2-3B","Collaboration Category":"Non-collaborative"}. This record does not assert independent reproduction.

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