Training Data Attribution on FAKEWIKI Obfuscate
44.4Recall@10SCORINGMODEL
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SCORINGMODELEvaluation Model Scope=Averaged across nine target LLMs2026.05 | 44.4 | 5.3 | 13.9 | |
| Best baselineSelection Strategy=Selected from eleven retrieval baselines separately for each model, Evaluation Model Scope=Averaged across nine target LLMs2026.05 | 39.1 | — | — | |
| STEERFUSEEvaluation Model Scope=Averaged across nine target LLMs2026.05 | 30.5 | — | — |