# Error Detection on KITTI DK (noise-induced test set)

Source: [Error Detection on KITTI DK (noise-induced test set)](<https://www.sota2.com/research/sota/error-detection-on-kitti-dk-noise-induced-test-set>)

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

Data updated: 2026-05-01T07:12:58.873Z

Evaluation of Error Detection performance on the noise-induced test subset of the KITTI DK dataset, measured using Precision, Recall, Accuracy, and F1 Score.

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 | Error Detection |
| Dataset (including any stated split) | KITTI DK:test^noise noise-induced (test) |
| Metric | Recall |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 15 of 15 |
| 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/error-detection-on-kitti-dk-noise-induced-test-set.md?metric=Recall>): Higher is better; unit: PERCENT.

- [Precision](<https://www.sota2.com/research/sota/error-detection-on-kitti-dk-noise-induced-test-set.md?metric=Precision>): Higher is better; unit: PERCENT.

- [Accuracy](<https://www.sota2.com/research/sota/error-detection-on-kitti-dk-noise-induced-test-set.md?metric=Accuracy>): Higher is better; unit: PERCENT.

- [F1 Score](<https://www.sota2.com/research/sota/error-detection-on-kitti-dk-noise-induced-test-set.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Recall | Result date | Report paper |
| --- | --- | --- | --- |
| DETR | 96 | Not recorded | 68fa7a081e209131cea80ac0 |
| Llama (FT-CoT) | 96 | Not recorded | 68fa7a081e209131cea80ac0 |
| AutoVDC (Llama (FT-CoT)) | 92 | Not recorded | 68fa7a081e209131cea80ac0 |
| Llama (FT) | 91 | Not recorded | 68fa7a081e209131cea80ac0 |
| AutoVDC (Llama (FT)) | 88 | Not recorded | 68fa7a081e209131cea80ac0 |
| Gemini Flash 2.0 | 84 | Not recorded | 68fa7a081e209131cea80ac0 |
| GPT 4.1 | 84 | Not recorded | 68fa7a081e209131cea80ac0 |
| AutoVDC (Gemini Flash 2.0) | 80 | Not recorded | 68fa7a081e209131cea80ac0 |
| AutoVDC (GPT 4.1) | 80 | Not recorded | 68fa7a081e209131cea80ac0 |
| ViP-LLaVA | 61 | Not recorded | 68fa7a081e209131cea80ac0 |
| DeepSeek-VL2 | 58 | Not recorded | 68fa7a081e209131cea80ac0 |
| AutoVDC (ViP-LLaVA) | 58 | Not recorded | 68fa7a081e209131cea80ac0 |
| AutoVDC (DeepSeek-VL2) | 55 | Not recorded | 68fa7a081e209131cea80ac0 |
| Llama | 47 | Not recorded | 68fa7a081e209131cea80ac0 |
| AutoVDC (Llama) | 45 | Not recorded | 68fa7a081e209131cea80ac0 |

### DETR

Recall: 96. Result ID: 69f4520c47f34ce924d9c27d.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa67ad1e209131cea17bce.md>)

[Implementation](<https://github.com/facebookresearch/detr>)

Recorded evaluation configuration: {"Evaluation Stage":"Error Proposals"}. This record does not assert independent reproduction.

### Llama (FT-CoT)

Recall: 96. Result ID: 69f4520c47f34ce924d9c28a.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"Error Validation","Mode":"Fine-Tuned with Chain-of-Thought"}. This record does not assert independent reproduction.

### AutoVDC (Llama (FT-CoT))

Recall: 92. Result ID: 69f4520c47f34ce924d9c28b.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"AutoVDC System","Backbone VLM":"Llama (Fine-Tuned with CoT)"}. This record does not assert independent reproduction.

### Llama (FT)

Recall: 91. Result ID: 69f4520c47f34ce924d9c288.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"Error Validation","Mode":"Fine-Tuned"}. This record does not assert independent reproduction.

### AutoVDC (Llama (FT))

Recall: 88. Result ID: 69f4520c47f34ce924d9c289.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"AutoVDC System","Backbone VLM":"Llama (Fine-Tuned)"}. This record does not assert independent reproduction.

### Gemini Flash 2.0

Recall: 84. Result ID: 69f4520c47f34ce924d9c280.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"Error Validation"}. This record does not assert independent reproduction.

### GPT 4.1

Recall: 84. Result ID: 69f4520c47f34ce924d9c282.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"Error Validation"}. This record does not assert independent reproduction.

### AutoVDC (Gemini Flash 2.0)

Recall: 80. Result ID: 69f4520c47f34ce924d9c281.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"AutoVDC System","Backbone VLM":"Gemini Flash 2.0"}. This record does not assert independent reproduction.

### AutoVDC (GPT 4.1)

Recall: 80. Result ID: 69f4520c47f34ce924d9c283.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"AutoVDC System","Backbone VLM":"GPT 4.1"}. This record does not assert independent reproduction.

### ViP-LLaVA

Recall: 61. Result ID: 69f4520c47f34ce924d9c284.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa73101e209131cea5a373.md>)

Recorded evaluation configuration: {"Evaluation Stage":"Error Validation"}. This record does not assert independent reproduction.

### DeepSeek-VL2

Recall: 58. Result ID: 69f4520c47f34ce924d9c27e.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa763a1e209131cea6bfb0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"Error Validation"}. This record does not assert independent reproduction.

### AutoVDC (ViP-LLaVA)

Recall: 58. Result ID: 69f4520c47f34ce924d9c285.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"AutoVDC System","Backbone VLM":"ViP-LLaVA"}. This record does not assert independent reproduction.

### AutoVDC (DeepSeek-VL2)

Recall: 55. Result ID: 69f4520c47f34ce924d9c27f.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"AutoVDC System","Backbone VLM":"DeepSeek-VL2"}. This record does not assert independent reproduction.

### Llama

Recall: 47. Result ID: 69f4520c47f34ce924d9c286.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa75d11e209131cea6a028.md>)

[Implementation](<https://github.com/meta-llama/llama-agentic-system>)

Recorded evaluation configuration: {"Evaluation Stage":"Error Validation","Mode":"Zero-shot"}. This record does not assert independent reproduction.

### AutoVDC (Llama)

Recall: 45. Result ID: 69f4520c47f34ce924d9c287.

[Reporting paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa7a081e209131cea80ac0.md>)

Recorded evaluation configuration: {"Evaluation Stage":"AutoVDC System","Backbone VLM":"Llama (Zero-shot)"}. This record does not assert independent reproduction.

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