# BEC Detection on BEC dataset

Source: [BEC Detection on BEC dataset](<https://www.sota2.com/research/sota/bec-detection-on-bec-dataset>)

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

Data updated: 2026-04-07T07:35:51.719Z

Evaluation of BEC Detection performance on the BEC dataset, measured using AUC, F1, Recall, and Precision.

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 | BEC Detection |
| Dataset (including any stated split) | BEC dataset |
| Metric | AUC |
| Unit | SCALAR |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [AUC](<https://www.sota2.com/research/sota/bec-detection-on-bec-dataset.md?metric=AUC>): Higher is better; unit: SCALAR.

- [F1 Score](<https://www.sota2.com/research/sota/bec-detection-on-bec-dataset.md?metric=F1%20Score>): Higher is better; unit: PERCENT.

- [Recall](<https://www.sota2.com/research/sota/bec-detection-on-bec-dataset.md?metric=Recall>): Higher is better; unit: PERCENT.

- [Precision](<https://www.sota2.com/research/sota/bec-detection-on-bec-dataset.md?metric=Precision>): Higher is better; unit: PERCENT.

## Recorded results

| Method | AUC | Result date | Report paper |
| --- | --- | --- | --- |
| DistilBERT (semantic) | 1 | Not recorded | 6927daf0d6b0dd3342930c0c |
| TF–IDF + LogReg (baseline) | 1 | Not recorded | 6927daf0d6b0dd3342930c0c |
| Char n-gram + SVM (baseline) | 1 | Not recorded | 6927daf0d6b0dd3342930c0c |
| CatBoost (forensic, +ψ) | 0.986 | Not recorded | 6927daf0d6b0dd3342930c0c |

### DistilBERT (semantic)

AUC: 1. Result ID: 69d4b30d47f34ce924bc2955.

[Reporting paper](<https://www.sota2.com/research/paper/6927daf0d6b0dd3342930c0c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927daf0d6b0dd3342930c0c.md>)

Recorded evaluation configuration: {"Latency (ms)":"7.403 (P95 10.873) GPU","Notes":"GPU inference"}. This record does not assert independent reproduction.

### TF–IDF + LogReg (baseline)

AUC: 1. Result ID: 69d4b30d47f34ce924bc2956.

[Reporting paper](<https://www.sota2.com/research/paper/6927daf0d6b0dd3342930c0c.md>)

Recorded evaluation configuration: {"Latency (ms)":"0.067 (P95 0.114) CPU","Expected Cost":"$2.52","Notes":"fastest baseline"}. This record does not assert independent reproduction.

### Char n-gram + SVM (baseline)

AUC: 1. Result ID: 69d4b30d47f34ce924bc2957.

[Reporting paper](<https://www.sota2.com/research/paper/6927daf0d6b0dd3342930c0c.md>)

Recorded evaluation configuration: {"Latency (ms)":"3.663 (P95 11.664) CPU","Expected Cost":"$0.36","Notes":"obfuscation-resistant"}. This record does not assert independent reproduction.

### CatBoost (forensic, +ψ)

AUC: 0.986. Result ID: 69d4b30d47f34ce924bc2958.

[Reporting paper](<https://www.sota2.com/research/paper/6927daf0d6b0dd3342930c0c.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6927daf0d6b0dd3342930c0c.md>)

Recorded evaluation configuration: {"Latency (ms)":"0.855 (P95 1.919) CPU","Expected Cost":"$13.57","Notes":"engineered features"}. This record does not assert independent reproduction.

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