# Descriptive Report Generation on Gut-VLM

Source: [Descriptive Report Generation on Gut-VLM](<https://www.sota2.com/research/sota/descriptive-report-generation-on-gut-vlm>)

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

Data updated: 2026-02-26T06:00:41.378Z

Evaluation of model performance on generating descriptive reports using the Gut-VLM dataset, measured by NLP metrics like ROUGE, BLEU, and METEOR, alongside specialized metrics like R-Sim and QAAS.

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 | Descriptive Report Generation |
| Dataset (including any stated split) | Gut-VLM |
| Metric | ROUGE-L |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 13 of 13 |
| 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

- [ROUGE-L](<https://www.sota2.com/research/sota/descriptive-report-generation-on-gut-vlm.md?metric=ROUGE-L>): Higher is better; unit: SCALAR.

- [BLEU](<https://www.sota2.com/research/sota/descriptive-report-generation-on-gut-vlm.md?metric=BLEU>): Higher is better; unit: SCALAR.

- [METEOR](<https://www.sota2.com/research/sota/descriptive-report-generation-on-gut-vlm.md?metric=METEOR>): Higher is better; unit: SCALAR.

- [R-Sim](<https://www.sota2.com/research/sota/descriptive-report-generation-on-gut-vlm.md?metric=R-Sim>): Higher is better; unit: SCALAR.

- [QAAS](<https://www.sota2.com/research/sota/descriptive-report-generation-on-gut-vlm.md?metric=QAAS>): Higher is better; unit: PERCENT.

## Recorded results

| Method | ROUGE-L | Result date | Report paper |
| --- | --- | --- | --- |
| LLaVa-1.6-7b | 0.89 | Not recorded | 68fa79961e209131cea7e68c |
| DeepSeek7b | 0.88 | Not recorded | 68fa79961e209131cea7e68c |
| Qwen7b | 0.88 | Not recorded | 68fa79961e209131cea7e68c |
| ChatGPT-4 Omni | 0.87 | Not recorded | 68fa79961e209131cea7e68c |
| MPlugOwl2b | 0.85 | Not recorded | 68fa79961e209131cea7e68c |
| DeepSeek7b | 0.55 | Not recorded | 68fa79961e209131cea7e68c |
| LLaVa-1.6-7b | 0.54 | Not recorded | 68fa79961e209131cea7e68c |
| Qwen7b | 0.54 | Not recorded | 68fa79961e209131cea7e68c |
| MPlugOwl2b | 0.5 | Not recorded | 68fa79961e209131cea7e68c |
| Qwen7b | 0.32 | Not recorded | 68fa79961e209131cea7e68c |
| DeepSeek7b | 0.29 | Not recorded | 68fa79961e209131cea7e68c |
| LLaVa-1.6-7b | 0.26 | Not recorded | 68fa79961e209131cea7e68c |
| MPlugOwl2b | 0.26 | Not recorded | 68fa79961e209131cea7e68c |

### LLaVa-1.6-7b

ROUGE-L: 0.89. Result ID: 698b54bb7a967eddb76db03b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"LLaVa-1.6-7b","Training Protocol":"Hallucination-aware finetuning"}. This record does not assert independent reproduction.

### DeepSeek7b

ROUGE-L: 0.88. Result ID: 698b54bb7a967eddb76db03e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-7b","Training Protocol":"Hallucination-aware finetuning"}. This record does not assert independent reproduction.

### Qwen7b

ROUGE-L: 0.88. Result ID: 698b54bb7a967eddb76db041.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"Qwen-7b","Training Protocol":"Hallucination-aware finetuning"}. This record does not assert independent reproduction.

### ChatGPT-4 Omni

ROUGE-L: 0.87. Result ID: 698b54bb7a967eddb76db038.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa727a1e209131cea56a1e.md>)

[Implementation](<https://github.com/openai/evals>)

Recorded evaluation configuration: {"Backbone":"ChatGPT-4","Training Protocol":"Proprietary/Original"}. This record does not assert independent reproduction.

### MPlugOwl2b

ROUGE-L: 0.85. Result ID: 698b54bb7a967eddb76db044.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"MPlugOwl-2b","Training Protocol":"Hallucination-aware finetuning"}. This record does not assert independent reproduction.

### DeepSeek7b

ROUGE-L: 0.55. Result ID: 698b54bb7a967eddb76db03d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-7b","Training Protocol":"Standard Finetuned"}. This record does not assert independent reproduction.

### LLaVa-1.6-7b

ROUGE-L: 0.54. Result ID: 698b54bb7a967eddb76db03a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"LLaVa-1.6-7b","Training Protocol":"Standard Finetuned"}. This record does not assert independent reproduction.

### Qwen7b

ROUGE-L: 0.54. Result ID: 698b54bb7a967eddb76db040.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"Qwen-7b","Training Protocol":"Standard Finetuned"}. This record does not assert independent reproduction.

### MPlugOwl2b

ROUGE-L: 0.5. Result ID: 698b54bb7a967eddb76db043.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"MPlugOwl-2b","Training Protocol":"Standard Finetuned"}. This record does not assert independent reproduction.

### Qwen7b

ROUGE-L: 0.32. Result ID: 698b54bb7a967eddb76db03f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"Qwen-7b","Training Protocol":"Pretrained"}. This record does not assert independent reproduction.

### DeepSeek7b

ROUGE-L: 0.29. Result ID: 698b54bb7a967eddb76db03c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"DeepSeek-7b","Training Protocol":"Pretrained"}. This record does not assert independent reproduction.

### LLaVa-1.6-7b

ROUGE-L: 0.26. Result ID: 698b54bb7a967eddb76db039.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"LLaVa-1.6-7b","Training Protocol":"Pretrained"}. This record does not assert independent reproduction.

### MPlugOwl2b

ROUGE-L: 0.26. Result ID: 698b54bb7a967eddb76db042.

[Reporting paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79961e209131cea7e68c.md>)

[Implementation](<https://github.com/bhattarailab/Hallucination-Aware-VLM>)

Recorded evaluation configuration: {"Backbone":"MPlugOwl-2b","Training Protocol":"Pretrained"}. This record does not assert independent reproduction.

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