# Wide Baseline Matching on ReasonMatch-Bench Overall (test)

Source: [Wide Baseline Matching on ReasonMatch-Bench Overall (test)](<https://www.sota2.com/research/sota/wide-baseline-matching-on-reasonmatch-bench-overall-test>)

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

Data updated: 2026-06-04T09:40:50.490Z

Evaluation of Wide Baseline Matching performance on the test split of the ReasonMatch-Bench Overall dataset, reporting F1, Precision, and Recall.

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 | Wide Baseline Matching |
| Dataset (including any stated split) | ReasonMatch-Bench Overall (test) |
| Metric | F1 Score |
| 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

- [F1 Score](<https://www.sota2.com/research/sota/wide-baseline-matching-on-reasonmatch-bench-overall-test.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Precision](<https://www.sota2.com/research/sota/wide-baseline-matching-on-reasonmatch-bench-overall-test.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/wide-baseline-matching-on-reasonmatch-bench-overall-test.md?metric=Recall>): Higher is better; unit: PERCENT.

## Recorded results

| Method | F1 Score | Result date | Report paper |
| --- | --- | --- | --- |
| Qwen3-VL-8B + DCRL | 70.5 | Not recorded | 6a2106a5c10484bac23f02fc |
| GPT-5-mini | 57.9 | Not recorded | 6a2106a5c10484bac23f02fc |
| GPT-5-Chat | 51.5 | Not recorded | 6a2106a5c10484bac23f02fc |
| Qwen3-VL-235B | 49.2 | Not recorded | 6a2106a5c10484bac23f02fc |
| Gemini-2.5-Pro | 42.8 | Not recorded | 6a2106a5c10484bac23f02fc |
| Claude-4.5-Sonnet | 41.7 | Not recorded | 6a2106a5c10484bac23f02fc |
| Claude-4-Sonnet | 34.8 | Not recorded | 6a2106a5c10484bac23f02fc |
| Claude-4.1-Opus | 33.7 | Not recorded | 6a2106a5c10484bac23f02fc |
| GPT-4o-241106 | 33.5 | Not recorded | 6a2106a5c10484bac23f02fc |
| Qwen3-VL-8B-Instruct | 27.5 | Not recorded | 6a2106a5c10484bac23f02fc |

### Qwen3-VL-8B + DCRL

F1 Score: 70.5. Result ID: 6a214814fbc373592c1d47d2.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

[Implementation](<https://github.com/aim-uofa/ReasonMatch>)

Recorded evaluation configuration: {"Backbone":"Qwen3-VL-8B","Method":"DCRL"}. This record does not assert independent reproduction.

### GPT-5-mini

F1 Score: 57.9. Result ID: 6a214814fbc373592c1d47c9.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

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

### GPT-5-Chat

F1 Score: 51.5. Result ID: 6a214814fbc373592c1d47ca.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

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

### Qwen3-VL-235B

F1 Score: 49.2. Result ID: 6a214814fbc373592c1d47d0.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Parameters":"235B"}. This record does not assert independent reproduction.

### Gemini-2.5-Pro

F1 Score: 42.8. Result ID: 6a214814fbc373592c1d47cc.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

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

### Claude-4.5-Sonnet

F1 Score: 41.7. Result ID: 6a214814fbc373592c1d47cd.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

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

### Claude-4-Sonnet

F1 Score: 34.8. Result ID: 6a214814fbc373592c1d47cf.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

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

### Claude-4.1-Opus

F1 Score: 33.7. Result ID: 6a214814fbc373592c1d47ce.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

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

### GPT-4o-241106

F1 Score: 33.5. Result ID: 6a214814fbc373592c1d47cb.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

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

### Qwen3-VL-8B-Instruct

F1 Score: 27.5. Result ID: 6a214814fbc373592c1d47d1.

[Reporting paper](<https://www.sota2.com/research/paper/6a2106a5c10484bac23f02fc.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927db54d6b0dd3342930f66.md>)

Recorded evaluation configuration: {"Parameters":"8B","Mode":"Instruct"}. This record does not assert independent reproduction.

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