# Visual Question Answering on MSVQA Underwater T=5 (test)

Source: [Visual Question Answering on MSVQA Underwater T=5 (test)](<https://www.sota2.com/research/sota/visual-question-answering-on-msvqa-underwater-t-5-test>)

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

Data updated: 2026-03-16T09:31:19.747Z

Evaluation of model performance on the test split of the MSVQA Underwater T=5 dataset for Visual Question Answering, reporting Average Performance (A-bar) and Last Performance (At).

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 | Visual Question Answering |
| Dataset (including any stated split) | MSVQA Underwater T=5 (test) |
| Metric | Last Performance (At) |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 10 of 10 |
| 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

- [Last Performance (At)](<https://www.sota2.com/research/sota/visual-question-answering-on-msvqa-underwater-t-5-test.md?metric=Last%20Performance%20(At)>): Higher is better; unit: PERCENT.

- [Average Performance (A-bar)](<https://www.sota2.com/research/sota/visual-question-answering-on-msvqa-underwater-t-5-test.md?metric=Average%20Performance%20(A-bar)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Last Performance (At) | Result date | Report paper |
| --- | --- | --- | --- |
| Joint | 84.27 | Not recorded | 69253816d6b0dd334292f46e |
| UNIFIER | 82.2 | Not recorded | 69253816d6b0dd334292f46e |
| QUAD | 79.62 | Not recorded | 69253816d6b0dd334292f46e |
| PODNet | 79.38 | Not recorded | 69253816d6b0dd334292f46e |
| VQACL | 78.6 | Not recorded | 69253816d6b0dd334292f46e |
| ER | 78.16 | Not recorded | 69253816d6b0dd334292f46e |
| Tailor | 76.97 | Not recorded | 69253816d6b0dd334292f46e |
| EWC | 76.14 | Not recorded | 69253816d6b0dd334292f46e |
| Finetune | 74.98 | Not recorded | 69253816d6b0dd334292f46e |
| Zero-shot | 19.3 | Not recorded | 69253816d6b0dd334292f46e |

### Joint

Last Performance (At): 84.27. Result ID: 69b7cde747f34ce9249fd17d.

[Reporting paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

Recorded evaluation configuration: {"Mem.":"✘","Number of tasks (T)":"5"}. This record does not assert independent reproduction.

### UNIFIER

Last Performance (At): 82.2. Result ID: 69b7cde747f34ce9249fd185.

[Reporting paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

Recorded evaluation configuration: {"Mem.":"✘","Number of tasks (T)":"5"}. This record does not assert independent reproduction.

### QUAD

Last Performance (At): 79.62. Result ID: 69b7cde747f34ce9249fd184.

[Reporting paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a311e209131cea81a6f.md>)

Recorded evaluation configuration: {"Mem.":"Questions","Number of tasks (T)":"5"}. This record does not assert independent reproduction.

### PODNet

Last Performance (At): 79.38. Result ID: 69b7cde747f34ce9249fd182.

[Reporting paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa686b1e209131cea1bffa.md>)

[Implementation](<https://www.sota2.com/github.com/arthurdouillard/incremental_learning.pytorch>)

Recorded evaluation configuration: {"Mem.":"Questions + Images","Number of tasks (T)":"5"}. This record does not assert independent reproduction.

### VQACL

Last Performance (At): 78.6. Result ID: 69b7cde747f34ce9249fd183.

[Reporting paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

Recorded evaluation configuration: {"Mem.":"Questions + Images","Number of tasks (T)":"5"}. This record does not assert independent reproduction.

### ER

Last Performance (At): 78.16. Result ID: 69b7cde747f34ce9249fd181.

[Reporting paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

Recorded evaluation configuration: {"Mem.":"Questions + Images","Number of tasks (T)":"5"}. This record does not assert independent reproduction.

### Tailor

Last Performance (At): 76.97. Result ID: 69b7cde747f34ce9249fd180.

[Reporting paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa723b1e209131cea55215.md>)

Recorded evaluation configuration: {"Mem.":"✘","Number of tasks (T)":"5"}. This record does not assert independent reproduction.

### EWC

Last Performance (At): 76.14. Result ID: 69b7cde747f34ce9249fd17f.

[Reporting paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa634f1e209131ce9fdfbe.md>)

Recorded evaluation configuration: {"Mem.":"✘","Number of tasks (T)":"5"}. This record does not assert independent reproduction.

### Finetune

Last Performance (At): 74.98. Result ID: 69b7cde747f34ce9249fd17e.

[Reporting paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

Recorded evaluation configuration: {"Mem.":"✘","Number of tasks (T)":"5"}. This record does not assert independent reproduction.

### Zero-shot

Last Performance (At): 19.3. Result ID: 69b7cde747f34ce9249fd17c.

[Reporting paper](<https://www.sota2.com/research/paper/69253816d6b0dd334292f46e.md>)

Recorded evaluation configuration: {"Mem.":"✘","Number of tasks (T)":"5"}. This record does not assert independent reproduction.

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