# Visual Reasoning and Calibration on Aggregate of 12 evaluation benchmarks Average

Source: [Visual Reasoning and Calibration on Aggregate of 12 evaluation benchmarks Average](<https://www.sota2.com/research/sota/visual-reasoning-and-calibration-on-aggregate-of-12-evaluation-benchmarks-average>)

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

Data updated: 2026-04-13T10:14:14.359Z

Aggregate evaluation across 12 distinct benchmarks focusing on visual reasoning capabilities and model calibration, reporting accuracy, AUROC, and Expected Calibration Error (ECE).

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 Reasoning and Calibration |
| Dataset (including any stated split) | Aggregate of 12 evaluation benchmarks Average |
| Metric | 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

- [Accuracy](<https://www.sota2.com/research/sota/visual-reasoning-and-calibration-on-aggregate-of-12-evaluation-benchmarks-average.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [AUROC](<https://www.sota2.com/research/sota/visual-reasoning-and-calibration-on-aggregate-of-12-evaluation-benchmarks-average.md?metric=AUROC>): Higher is better; unit: PERCENT.

- [ECE](<https://www.sota2.com/research/sota/visual-reasoning-and-calibration-on-aggregate-of-12-evaluation-benchmarks-average.md?metric=ECE>): Lower is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| VL-Calibration | 80.3 | Not recorded | 69dc78a77d0172140ec78c5a |
| VL-Calibration | 68.9 | Not recorded | 69dc78a77d0172140ec78c5a |
| RLCR | 65.6 | Not recorded | 69dc78a77d0172140ec78c5a |
| P(True) | 65.2 | Not recorded | 69dc78a77d0172140ec78c5a |

### VL-Calibration

Accuracy: 80.3. Result ID: 69dcc1da47f34ce924c43be5.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a77d0172140ec78c5a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69dc78a77d0172140ec78c5a.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-30B"}. This record does not assert independent reproduction.

### VL-Calibration

Accuracy: 68.9. Result ID: 69dcc1da47f34ce924c43be7.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a77d0172140ec78c5a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69dc78a77d0172140ec78c5a.md>)

Recorded evaluation configuration: {"Model":"InternVL3.5-4B"}. This record does not assert independent reproduction.

### RLCR

Accuracy: 65.6. Result ID: 69dcc1da47f34ce924c43be6.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a77d0172140ec78c5a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a241e209131cea81436.md>)

Recorded evaluation configuration: {"Model":"InternVL3.5-4B"}. This record does not assert independent reproduction.

### P(True)

Accuracy: 65.2. Result ID: 69dcc1da47f34ce924c43be4.

[Reporting paper](<https://www.sota2.com/research/paper/69dc78a77d0172140ec78c5a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa6df71e209131cea3d034.md>)

Recorded evaluation configuration: {"Model":"Qwen3-VL-30B"}. This record does not assert independent reproduction.

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