# Model Attribution on Section 5.2 corpus 6 AIs x 3 domains (sample-level leave-one-out) Source: [Model Attribution on Section 5.2 corpus 6 AIs x 3 domains (sample-level leave-one-out)]() Collection index: [llms.txt]() Data updated: 2026-06-16T06:51:38.910Z Evaluation of Model Attribution performance using a sample-level leave-one-out strategy across a corpus involving 6 AIs and 3 domains, reporting 6-way LOO accuracy and confidence intervals. 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 | Model Attribution | | Dataset (including any stated split) | Section 5.2 corpus 6 AIs x 3 domains (sample-level leave-one-out) | | Metric | 6-way LOO 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 - [6-way LOO Accuracy](): Higher is better; unit: PERCENT. - [6-way LOO Accuracy 95% CI Lower Bound](): Higher is better; unit: SCALAR. ## Recorded results | Method | 6-way LOO Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Reference-anchored (patch\_mean, ssim) | 51.3 | Not recorded | 6a30d88ed33dfb23e783cbd8 | | PRISM-style (DINOv2 CLS, blind) | 37.3 | Not recorded | 6a30d88ed33dfb23e783cbd8 | ### Reference-anchored (patch\_mean, ssim) 6-way LOO Accuracy: 51.3. Result ID: 6a30f278fbc373592c2c8f80. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Reference":"yes","Features":"patch\_mean, ssim","Evaluation protocol":"sample-level leave-one-out nearest-centroid","Resampling":"2,000-iteration cluster bootstrap"}. This record does not assert independent reproduction. ### PRISM-style (DINOv2 CLS, blind) 6-way LOO Accuracy: 37.3. Result ID: 6a30f278fbc373592c2c8f81. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Reference":"no","Backbone":"DINOv2 ViT-B/14 CLS","Evaluation protocol":"sample-level leave-one-out nearest-centroid","Resampling":"2,000-iteration cluster bootstrap"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.