# Visual-to-Symbolic Analytical Solution Inference on ViSA-Bench VLM setting

Source: [Visual-to-Symbolic Analytical Solution Inference on ViSA-Bench VLM setting](<https://www.sota2.com/research/sota/visual-to-symbolic-analytical-solution-inference-on-visa-bench-vlm-setting>)

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

Data updated: 2026-04-13T08:09:46.905Z

Evaluation of model performance on the ViSA-Bench dataset (VLM setting) focusing on Visual-to-Symbolic Analytical Solution Inference, measured by Overall, Character, Structure, and Numerical accuracy.

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-to-Symbolic Analytical Solution Inference |
| Dataset (including any stated split) | ViSA-Bench VLM setting |
| Metric | Overall Accuracy |
| 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

- [Overall Accuracy](<https://www.sota2.com/research/sota/visual-to-symbolic-analytical-solution-inference-on-visa-bench-vlm-setting.md?metric=Overall%20Accuracy>): Higher is better; unit: PERCENT.

- [Character Accuracy](<https://www.sota2.com/research/sota/visual-to-symbolic-analytical-solution-inference-on-visa-bench-vlm-setting.md?metric=Character%20Accuracy>): Higher is better; unit: PERCENT.

- [Structure Accuracy](<https://www.sota2.com/research/sota/visual-to-symbolic-analytical-solution-inference-on-visa-bench-vlm-setting.md?metric=Structure%20Accuracy>): Higher is better; unit: PERCENT.

- [Numerical Accuracy](<https://www.sota2.com/research/sota/visual-to-symbolic-analytical-solution-inference-on-visa-bench-vlm-setting.md?metric=Numerical%20Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Overall Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| ViSA-R2 | 51.2 | Not recorded | 69dc78997d0172140ec78bb4 |
| GPT-5.2 | 47.4 | Not recorded | 69dc78997d0172140ec78bb4 |
| Claude-Haiku-4-5 | 31.6 | Not recorded | 69dc78997d0172140ec78bb4 |
| Grok-4-1-Fast | 24.2 | Not recorded | 69dc78997d0172140ec78bb4 |

### ViSA-R2

Overall Accuracy: 51.2. Result ID: 69dca44147f34ce924c379b9.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78997d0172140ec78bb4.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69dc78997d0172140ec78bb4.md>)

Recorded evaluation configuration: {"Input Mode":"VLM"}. This record does not assert independent reproduction.

### GPT-5.2

Overall Accuracy: 47.4. Result ID: 69dca44147f34ce924c379b6.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78997d0172140ec78bb4.md>)

Recorded evaluation configuration: {"Input Mode":"VLM"}. This record does not assert independent reproduction.

### Claude-Haiku-4-5

Overall Accuracy: 31.6. Result ID: 69dca44147f34ce924c379b8.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78997d0172140ec78bb4.md>)

Recorded evaluation configuration: {"Input Mode":"VLM"}. This record does not assert independent reproduction.

### Grok-4-1-Fast

Overall Accuracy: 24.2. Result ID: 69dca44147f34ce924c379b7.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78997d0172140ec78bb4.md>)

Recorded evaluation configuration: {"Input Mode":"VLM"}. This record does not assert independent reproduction.

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