# Selective Prediction on To See or To Please Diagnostic

Source: [Selective Prediction on To See or To Please Diagnostic](<https://www.sota2.com/research/sota/selective-prediction-on-to-see-or-to-please-diagnostic>)

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

Data updated: 2026-03-20T07:03:03.009Z

Evaluation of Selective Prediction performance on the 'To See or To Please Diagnostic' dataset, using metrics such as Baseline accuracy, Accuracy at 50% Coverage, and Delta (Δ).

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 | Selective Prediction |
| Dataset (including any stated split) | To See or To Please Diagnostic |
| Metric | Baseline Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 7 of 7 |
| 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

- [Baseline Accuracy](<https://www.sota2.com/research/sota/selective-prediction-on-to-see-or-to-please-diagnostic.md?metric=Baseline%20Accuracy>): Higher is better; unit: PERCENT.

- [Accuracy @ 50% Coverage](<https://www.sota2.com/research/sota/selective-prediction-on-to-see-or-to-please-diagnostic.md?metric=Accuracy%20%40%2050%25%20Coverage>): Higher is better; unit: PERCENT.

- [Delta (Δ)](<https://www.sota2.com/research/sota/selective-prediction-on-to-see-or-to-please-diagnostic.md?metric=Delta%20(%CE%94)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Baseline Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen2.5-VL-72B | 73.5 | Not recorded | 69bcd46d0636e5bf4afc28b1 |
| Qwen2.5-VL-7B | 72.1 | Not recorded | 69bcd46d0636e5bf4afc28b1 |
| Molmo2-4B | 71.3 | Not recorded | 69bcd46d0636e5bf4afc28b1 |
| Llama-3.2-11B | 69.9 | Not recorded | 69bcd46d0636e5bf4afc28b1 |
| Phi-3.5-Vision | 69 | Not recorded | 69bcd46d0636e5bf4afc28b1 |
| LLaVA-NeXT-7B | 68.1 | Not recorded | 69bcd46d0636e5bf4afc28b1 |
| Pixtral-12B | 66.9 | Not recorded | 69bcd46d0636e5bf4afc28b1 |

### Qwen2.5-VL-72B

Baseline Accuracy: 73.5. Result ID: 69bcf11847f34ce924a6b0e9.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

Recorded evaluation configuration: {"Strategy":"Diagnostic-Guided Selective Prediction","Coverage":"50%"}. This record does not assert independent reproduction.

### Qwen2.5-VL-7B

Baseline Accuracy: 72.1. Result ID: 69bcf11847f34ce924a6b0e6.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

Recorded evaluation configuration: {"Strategy":"Diagnostic-Guided Selective Prediction","Coverage":"50%"}. This record does not assert independent reproduction.

### Molmo2-4B

Baseline Accuracy: 71.3. Result ID: 69bcf11847f34ce924a6b0e8.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

Recorded evaluation configuration: {"Strategy":"Diagnostic-Guided Selective Prediction","Coverage":"50%"}. This record does not assert independent reproduction.

### Llama-3.2-11B

Baseline Accuracy: 69.9. Result ID: 69bcf11847f34ce924a6b0eb.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

Recorded evaluation configuration: {"Strategy":"Diagnostic-Guided Selective Prediction","Coverage":"50%"}. This record does not assert independent reproduction.

### Phi-3.5-Vision

Baseline Accuracy: 69. Result ID: 69bcf11847f34ce924a6b0ea.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

Recorded evaluation configuration: {"Strategy":"Diagnostic-Guided Selective Prediction","Coverage":"50%"}. This record does not assert independent reproduction.

### LLaVA-NeXT-7B

Baseline Accuracy: 68.1. Result ID: 69bcf11847f34ce924a6b0e7.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

Recorded evaluation configuration: {"Strategy":"Diagnostic-Guided Selective Prediction","Coverage":"50%"}. This record does not assert independent reproduction.

### Pixtral-12B

Baseline Accuracy: 66.9. Result ID: 69bcf11847f34ce924a6b0ec.

[Reporting paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69bcd46d0636e5bf4afc28b1.md>)

Recorded evaluation configuration: {"Strategy":"Diagnostic-Guided Selective Prediction","Coverage":"50%"}. This record does not assert independent reproduction.

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