# Online auditing on Who&When

Source: [Online auditing on Who&When](<https://www.sota2.com/research/sota/online-auditing-on-who-and-when>)

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

Data updated: 2026-05-12T09:30:54.189Z

Evaluation of online auditing performance on the Who&When dataset, measured using Step-Acc, Agent-Acc, and ASS metrics.

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 | Online auditing |
| Dataset (including any stated split) | Who&When |
| Metric | Step Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 8 of 8 |
| 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

- [Step Accuracy](<https://www.sota2.com/research/sota/online-auditing-on-who-and-when.md?metric=Step%20Accuracy>): Higher is better; unit: PERCENT.

- [Agent Accuracy](<https://www.sota2.com/research/sota/online-auditing-on-who-and-when.md?metric=Agent%20Accuracy>): Higher is better; unit: PERCENT.

- [ASS](<https://www.sota2.com/research/sota/online-auditing-on-who-and-when.md?metric=ASS>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Step Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| AgentForesight-7B | 57.69 | Not recorded | 6a02b436dbc836b87d82484c |
| GPT-4.1 | 38.1 | Not recorded | 6a02b436dbc836b87d82484c |
| DeepSeek-V4-Flash | 37.21 | Not recorded | 6a02b436dbc836b87d82484c |
| Qwen2.5-7B-Instruct | 36.59 | Not recorded | 6a02b436dbc836b87d82484c |
| Gemini-3-Flash | 32.56 | Not recorded | 6a02b436dbc836b87d82484c |
| Qwen3-8B | 29.41 | Not recorded | 6a02b436dbc836b87d82484c |
| Llama3.2-3B | 28.57 | Not recorded | 6a02b436dbc836b87d82484c |
| Gemma3-4B | 6.98 | Not recorded | 6a02b436dbc836b87d82484c |

### AgentForesight-7B

Step Accuracy: 57.69. Result ID: 6a02f2db47f34ce924e602fd.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b436dbc836b87d82484c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6a02b436dbc836b87d82484c.md>)

[Implementation](<https://github.com/ZBox1005/AgentForesight>)

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

### GPT-4.1

Step Accuracy: 38.1. Result ID: 6a02f2db47f34ce924e602fa.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b436dbc836b87d82484c.md>)

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

### DeepSeek-V4-Flash

Step Accuracy: 37.21. Result ID: 6a02f2db47f34ce924e602fc.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b436dbc836b87d82484c.md>)

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

### Qwen2.5-7B-Instruct

Step Accuracy: 36.59. Result ID: 6a02f2db47f34ce924e602f8.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b436dbc836b87d82484c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

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

### Gemini-3-Flash

Step Accuracy: 32.56. Result ID: 6a02f2db47f34ce924e602fb.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b436dbc836b87d82484c.md>)

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

### Qwen3-8B

Step Accuracy: 29.41. Result ID: 6a02f2db47f34ce924e602f9.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b436dbc836b87d82484c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa788e1e209131cea79116.md>)

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

### Llama3.2-3B

Step Accuracy: 28.57. Result ID: 6a02f2db47f34ce924e602f6.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b436dbc836b87d82484c.md>)

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

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

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

### Gemma3-4B

Step Accuracy: 6.98. Result ID: 6a02f2db47f34ce924e602f7.

[Reporting paper](<https://www.sota2.com/research/paper/6a02b436dbc836b87d82484c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa779f1e209131cea746cd.md>)

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

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