# LLM Detection Interpretability on M4 (test) Source: [LLM Detection Interpretability on M4 (test)]() Collection index: [llms.txt]() Data updated: 2026-05-06T07:25:09.534Z Evaluation of the interpretability of LLM detection models using the test split of the M4 dataset, measured by the accuracy of human judgments. 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 | LLM Detection Interpretability | | Dataset (including any stated split) | M4 (test) | | Metric | Accuracy (Human Judgements) | | Unit | PERCENT | | 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 - [Accuracy (Human Judgements)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (Human Judgements) | Result date | Report paper | | --- | --- | --- | --- | | ExaGPT | 61.5 | Not recorded | 68fa76fe1e209131cea704c7 | | LR-GLTR | 57.3 | Not recorded | 68fa76fe1e209131cea704c7 | | DNA-GPT | 53.1 | Not recorded | 68fa76fe1e209131cea704c7 | | RoBERTa | 47.9 | Not recorded | 68fa76fe1e209131cea704c7 | ### ExaGPT Accuracy (Human Judgements): 61.5. Result ID: 69faec5e47f34ce924dd7494. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### LR-GLTR Accuracy (Human Judgements): 57.3. Result ID: 69faec5e47f34ce924dd7492. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### DNA-GPT Accuracy (Human Judgements): 53.1. Result ID: 69faec5e47f34ce924dd7493. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### RoBERTa Accuracy (Human Judgements): 47.9. Result ID: 69faec5e47f34ce924dd7491. [Reporting paper]() [Method origin paper]() 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.