# Poisoned Sample Detection on NQ (IID) Source: [Poisoned Sample Detection on NQ (IID)]() Collection index: [llms.txt]() Data updated: 2026-03-04T09:48:26.384Z Evaluation of model performance for poisoned sample detection using the NQ (IID) dataset, measured by Recall and F1 score. 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 | Poisoned Sample Detection | | Dataset (including any stated split) | NQ (IID) | | Metric | Recall | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | 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 - [Recall](): Higher is better; unit: PERCENT. - [F1 Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Recall | Result date | Report paper | | --- | --- | --- | --- | | ProtegoFed | 100 | Not recorded | 69a66bc82d603e6de3e3cb60 | | ProtegoFed | 100 | Not recorded | 69a66bc82d603e6de3e3cb60 | | ProtegoFed | 100 | Not recorded | 69a66bc82d603e6de3e3cb60 | | ProtegoFed | 99.4 | Not recorded | 69a66bc82d603e6de3e3cb60 | | GraCeFul | 95.6 | Not recorded | 69a66bc82d603e6de3e3cb60 | | GraCeFul | 94.8 | Not recorded | 69a66bc82d603e6de3e3cb60 | | Onion | 88.4 | Not recorded | 69a66bc82d603e6de3e3cb60 | | GraCeFul | 87.8 | Not recorded | 69a66bc82d603e6de3e3cb60 | | GraCeFul | 82.4 | Not recorded | 69a66bc82d603e6de3e3cb60 | | CUBE | 0 | Not recorded | 69a66bc82d603e6de3e3cb60 | | Onion | 0 | Not recorded | 69a66bc82d603e6de3e3cb60 | | CUBE | 0 | Not recorded | 69a66bc82d603e6de3e3cb60 | | Onion | 0 | Not recorded | 69a66bc82d603e6de3e3cb60 | | CUBE | 0 | Not recorded | 69a66bc82d603e6de3e3cb60 | | Onion | 0 | Not recorded | 69a66bc82d603e6de3e3cb60 | | CUBE | 0 | Not recorded | 69a66bc82d603e6de3e3cb60 | ### ProtegoFed Recall: 100. Result ID: 69a7ffdf6dbb7075b94d41bb. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Poison Method":"Badnets"}. This record does not assert independent reproduction. ### ProtegoFed Recall: 100. Result ID: 69a7ffdf6dbb7075b94d41bf. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Poison Method":"Addsent"}. This record does not assert independent reproduction. ### ProtegoFed Recall: 100. Result ID: 69a7ffdf6dbb7075b94d41c7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Poison Method":"StyleBkd"}. This record does not assert independent reproduction. ### ProtegoFed Recall: 99.4. Result ID: 69a7ffdf6dbb7075b94d41c3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Poison Method":"CBA"}. This record does not assert independent reproduction. ### GraCeFul Recall: 95.6. Result ID: 69a7ffdf6dbb7075b94d41b9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"Badnets"}. This record does not assert independent reproduction. ### GraCeFul Recall: 94.8. Result ID: 69a7ffdf6dbb7075b94d41bd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"Addsent"}. This record does not assert independent reproduction. ### Onion Recall: 88.4. Result ID: 69a7ffdf6dbb7075b94d41c6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"StyleBkd"}. This record does not assert independent reproduction. ### GraCeFul Recall: 87.8. Result ID: 69a7ffdf6dbb7075b94d41c1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"CBA"}. This record does not assert independent reproduction. ### GraCeFul Recall: 82.4. Result ID: 69a7ffdf6dbb7075b94d41c5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"StyleBkd"}. This record does not assert independent reproduction. ### CUBE Recall: 0. Result ID: 69a7ffdf6dbb7075b94d41b8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"Badnets"}. This record does not assert independent reproduction. ### Onion Recall: 0. Result ID: 69a7ffdf6dbb7075b94d41ba. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"Badnets"}. This record does not assert independent reproduction. ### CUBE Recall: 0. Result ID: 69a7ffdf6dbb7075b94d41bc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"Addsent"}. This record does not assert independent reproduction. ### Onion Recall: 0. Result ID: 69a7ffdf6dbb7075b94d41be. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"Addsent"}. This record does not assert independent reproduction. ### CUBE Recall: 0. Result ID: 69a7ffdf6dbb7075b94d41c0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"CBA"}. This record does not assert independent reproduction. ### Onion Recall: 0. Result ID: 69a7ffdf6dbb7075b94d41c2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"CBA"}. This record does not assert independent reproduction. ### CUBE Recall: 0. Result ID: 69a7ffdf6dbb7075b94d41c4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Poison Method":"StyleBkd"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.