# Dataset Diversity and Coverage Evaluation on AgentHazard full

Source: [Dataset Diversity and Coverage Evaluation on AgentHazard full](<https://www.sota2.com/research/sota/dataset-diversity-and-coverage-evaluation-on-agenthazard-full>)

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

Data updated: 2026-05-28T09:01:56.976Z

Evaluation of model performance on the AgentHazard full dataset, focusing on diversity and coverage metrics such as Goal-text entropy, CLIP pairwise distance, and Action coverage.

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 | Dataset Diversity and Coverage Evaluation |
| Dataset (including any stated split) | AgentHazard full |
| Metric | Goal-text Entropy |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 1 of 1 |
| 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

- [Goal-text Entropy](<https://www.sota2.com/research/sota/dataset-diversity-and-coverage-evaluation-on-agenthazard-full.md?metric=Goal-text%20Entropy>): Higher is better; unit: SCALAR.

- [CLIP Pairwise Distance (per-base)](<https://www.sota2.com/research/sota/dataset-diversity-and-coverage-evaluation-on-agenthazard-full.md?metric=CLIP%20Pairwise%20Distance%20(per-base)>): Higher is better; unit: SCALAR.

- [Action Coverage](<https://www.sota2.com/research/sota/dataset-diversity-and-coverage-evaluation-on-agenthazard-full.md?metric=Action%20Coverage>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Goal-text Entropy | Result date | Report paper |
| --- | --- | --- | --- |
| AgentHazard (full) | 0.93 | Not recorded | 6a17cc01dbd7150506f8123a |

### AgentHazard (full)

Goal-text Entropy: 0.93. Result ID: 6a18032ffbc373592c11e240.

[Reporting paper](<https://www.sota2.com/research/paper/6a17cc01dbd7150506f8123a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa79d51e209131cea7f988.md>)

[Implementation](<https://github.com/Zsbyqx20/AgentHazard>)

Recorded evaluation configuration: {"n":"2,520"}. This record does not assert independent reproduction.

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