# Region Understanding on ViP-Bench

Source: [Region Understanding on ViP-Bench](<https://www.sota2.com/research/sota/region-understanding-on-vip-bench>)

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

Data updated: 2026-02-26T06:01:45.035Z

Evaluation of model performance on the ViP-Bench dataset focusing on Region Understanding capabilities, measured across sub-tasks including Recognition, OCR, Knowledge, Math, Relationship, and Language Generation.

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 | Region Understanding |
| Dataset (including any stated split) | ViP-Bench |
| Metric | Recognition Score |
| 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

- [Recognition Score](<https://www.sota2.com/research/sota/region-understanding-on-vip-bench.md?metric=Recognition%20Score>): Higher is better; unit: PERCENT.

- [OCR Score](<https://www.sota2.com/research/sota/region-understanding-on-vip-bench.md?metric=OCR%20Score>): Higher is better; unit: PERCENT.

- [Knowledge Score](<https://www.sota2.com/research/sota/region-understanding-on-vip-bench.md?metric=Knowledge%20Score>): Higher is better; unit: PERCENT.

- [Math Score](<https://www.sota2.com/research/sota/region-understanding-on-vip-bench.md?metric=Math%20Score>): Higher is better; unit: PERCENT.

- [Relationship Score](<https://www.sota2.com/research/sota/region-understanding-on-vip-bench.md?metric=Relationship%20Score>): Higher is better; unit: PERCENT.

- [Language Generation Score](<https://www.sota2.com/research/sota/region-understanding-on-vip-bench.md?metric=Language%20Generation%20Score>): Higher is better; unit: PERCENT.

- [Overall Score](<https://www.sota2.com/research/sota/region-understanding-on-vip-bench.md?metric=Overall%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Recognition Score | Result date | Report paper |
| --- | --- | --- | --- |
| Shikra-7B | 0.402 | Not recorded | 68fa794e1e209131cea7d037 |
| InstructBLIP | 0.369 | Not recorded | 68fa794e1e209131cea7d037 |
| GPT4RoI-7B | 0.356 | Not recorded | 68fa794e1e209131cea7d037 |
| Kosmos-2 | 0.295 | Not recorded | 68fa794e1e209131cea7d037 |

### Shikra-7B

Recognition Score: 0.402. Result ID: 6995d9003badfaffe6df329b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa794e1e209131cea7d037.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fe61e209131cea47fbc.md>)

[Implementation](<https://github.com/shikras/shikra>)

Recorded evaluation configuration: {"Reference Format":"textual coordinates","Concurrent work":true}. This record does not assert independent reproduction.

### InstructBLIP

Recognition Score: 0.369. Result ID: 6995d9003badfaffe6df329a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa794e1e209131cea7d037.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6fb71e209131cea47021.md>)

[Implementation](<https://github.com/salesforce/LAVIS/tree/main/projects/instructblip>)

Recorded evaluation configuration: {"Reference Format":"pure language reference"}. This record does not assert independent reproduction.

### GPT4RoI-7B

Recognition Score: 0.356. Result ID: 6995d9003badfaffe6df329c.

[Reporting paper](<https://www.sota2.com/research/paper/68fa794e1e209131cea7d037.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa794e1e209131cea7d037.md>)

[Implementation](<https://github.com/jshilong/GPT4RoI>)

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

### Kosmos-2

Recognition Score: 0.295. Result ID: 6995d9003badfaffe6df3299.

[Reporting paper](<https://www.sota2.com/research/paper/68fa794e1e209131cea7d037.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa700a1e209131cea48c7f.md>)

[Implementation](<https://aka.ms/kosmos-2>)

Recorded evaluation configuration: {"Reference Format":"extra position tokens","Concurrent work":true}. This record does not assert independent reproduction.

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