# Infrastructure improvement proposal on BDD100K (test)

Source: [Infrastructure improvement proposal on BDD100K (test)](<https://www.sota2.com/research/sota/infrastructure-improvement-proposal-on-bdd100k-test>)

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

Data updated: 2026-03-09T07:51:26.411Z

Evaluation of model performance on generating or assessing infrastructure improvement proposals using the test split of the BDD100K dataset, reporting standard classification metrics like Recall, Precision, F1-Score, and 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 | Infrastructure improvement proposal |
| Dataset (including any stated split) | BDD100K (test) |
| Metric | Recall |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 9 of 9 |
| 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

- [Recall](<https://www.sota2.com/research/sota/infrastructure-improvement-proposal-on-bdd100k-test.md?metric=Recall>): Higher is better; unit: PERCENT.

- [Precision](<https://www.sota2.com/research/sota/infrastructure-improvement-proposal-on-bdd100k-test.md?metric=Precision>): Higher is better; unit: PERCENT.

- [F1-Score](<https://www.sota2.com/research/sota/infrastructure-improvement-proposal-on-bdd100k-test.md?metric=F1-Score>): Higher is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/infrastructure-improvement-proposal-on-bdd100k-test.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Recall | Result date | Report paper |
| --- | --- | --- | --- |
| OD-RASE (Long-CLIP) | 69.17 | Not recorded | 69ae53da2d603e6de3e4281a |
| OD-RASE (CLIP) | 66.86 | Not recorded | 69ae53da2d603e6de3e4281a |
| OD-RASE (ViT-B) | 62.21 | Not recorded | 69ae53da2d603e6de3e4281a |
| OD-RASE (ResNet-50) | 59.44 | Not recorded | 69ae53da2d603e6de3e4281a |
| Phi-3 | 51.35 | Not recorded | 69ae53da2d603e6de3e4281a |
| GPT-4o | 50.73 | Not recorded | 69ae53da2d603e6de3e4281a |
| LLaVA-1.5 | 46.71 | Not recorded | 69ae53da2d603e6de3e4281a |
| Qwen2-VL | 27.88 | Not recorded | 69ae53da2d603e6de3e4281a |
| InternVL2 | 25.11 | Not recorded | 69ae53da2d603e6de3e4281a |

### OD-RASE (Long-CLIP)

Recall: 69.17. Result ID: 69ae7bee6dbb7075b9565e90.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

Recorded evaluation configuration: {"Vision Encoder":"Long-CLIP","Text Encoder":"RoBERTa-Base","Zero-shot":"true"}. This record does not assert independent reproduction.

### OD-RASE (CLIP)

Recall: 66.86. Result ID: 69ae7bee6dbb7075b9565e8f.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

Recorded evaluation configuration: {"Vision Encoder":"CLIP","Text Encoder":"RoBERTa-Base","Zero-shot":"true"}. This record does not assert independent reproduction.

### OD-RASE (ViT-B)

Recall: 62.21. Result ID: 69ae7bee6dbb7075b9565e8e.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

Recorded evaluation configuration: {"Vision Encoder":"ViT-B","Text Encoder":"RoBERTa-Base","Zero-shot":"true"}. This record does not assert independent reproduction.

### OD-RASE (ResNet-50)

Recall: 59.44. Result ID: 69ae7bee6dbb7075b9565e8d.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

Recorded evaluation configuration: {"Vision Encoder":"ResNet-50","Text Encoder":"RoBERTa-Base","Zero-shot":"true"}. This record does not assert independent reproduction.

### Phi-3

Recall: 51.35. Result ID: 69ae7bee6dbb7075b9565e94.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa74ae1e209131cea632e1.md>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

### GPT-4o

Recall: 50.73. Result ID: 69ae7bee6dbb7075b9565e91.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75681e209131cea67a98.md>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

### LLaVA-1.5

Recall: 46.71. Result ID: 69ae7bee6dbb7075b9565e92.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa717d1e209131cea51970.md>)

[Implementation](<https://llava-vl.github.io/>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

### Qwen2-VL

Recall: 27.88. Result ID: 69ae7bee6dbb7075b9565e93.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75151e209131cea65976.md>)

[Implementation](<https://github.com/QwenLM/Qwen2-VL>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

### InternVL2

Recall: 25.11. Result ID: 69ae7bee6dbb7075b9565e95.

[Reporting paper](<https://www.sota2.com/research/paper/69ae53da2d603e6de3e4281a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa71de1e209131cea5370d.md>)

[Implementation](<https://github.com/OpenGVLab/InternVL>)

Recorded evaluation configuration: {"Zero-shot":"true"}. This record does not assert independent reproduction.

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