# Behavioral Attribution on Simulated Threat Actor Campaigns (temporal leave-one-out)

Source: [Behavioral Attribution on Simulated Threat Actor Campaigns (temporal leave-one-out)](<https://www.sota2.com/research/sota/behavioral-attribution-on-simulated-threat-actor-campaigns-temporal-leave-one-out>)

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

Data updated: 2026-04-28T10:35:30.493Z

Evaluation of Behavioral Attribution performance on the Simulated Threat Actor Campaigns dataset using a temporal leave-one-out evaluation strategy. Metrics include Overall accuracy, Mean confidence, and High-conf. 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 | Behavioral Attribution |
| Dataset (including any stated split) | Simulated Threat Actor Campaigns (temporal leave-one-out) |
| Metric | Overall Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| 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

- [Overall Accuracy](<https://www.sota2.com/research/sota/behavioral-attribution-on-simulated-threat-actor-campaigns-temporal-leave-one-out.md?metric=Overall%20Accuracy>): Higher is better; unit: PERCENT.

- [Mean Confidence](<https://www.sota2.com/research/sota/behavioral-attribution-on-simulated-threat-actor-campaigns-temporal-leave-one-out.md?metric=Mean%20Confidence>): Higher is better; unit: SCALAR.

- [High-Confidence Accuracy](<https://www.sota2.com/research/sota/behavioral-attribution-on-simulated-threat-actor-campaigns-temporal-leave-one-out.md?metric=High-Confidence%20Accuracy>): Higher is better; unit: PERCENT.

- [Fingerprint Separation (ΔS)](<https://www.sota2.com/research/sota/behavioral-attribution-on-simulated-threat-actor-campaigns-temporal-leave-one-out.md?metric=Fingerprint%20Separation%20(%CE%94S)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | Overall Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Baseline | 43.2 | Not recorded | 69f03f5c55a3c32af89fa3b3 |
| ARCANE | 30.7 | Not recorded | 69f03f5c55a3c32af89fa3b3 |

### Baseline

Overall Accuracy: 43.2. Result ID: 69f08d5f47f34ce924d73c26.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f5c55a3c32af89fa3b3.md>)

Recorded evaluation configuration: {"Methodology":"per-campaign nearest-neighbour","Evaluation Protocol":"temporal leave-one-out"}. This record does not assert independent reproduction.

### ARCANE

Overall Accuracy: 30.7. Result ID: 69f08d5f47f34ce924d73c25.

[Reporting paper](<https://www.sota2.com/research/paper/69f03f5c55a3c32af89fa3b3.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69f03f5c55a3c32af89fa3b3.md>)

Recorded evaluation configuration: {"Methodology":"longitudinal Bayesian accumulation","Evaluation Protocol":"temporal leave-one-out"}. This record does not assert independent reproduction.

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