# Adversarial Detection on AG News Source: [Adversarial Detection on AG News]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:43:16.613Z Evaluation of model performance in detecting adversarial manipulations applied to text samples from the AG News dataset. 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 | Adversarial Detection | | Dataset (including any stated split) | AG News | | Metric | F1 Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 18 of 18 | | 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 - [F1 Score](): Higher is better; unit: PERCENT. - [Adversarial Recall](): Higher is better; unit: PERCENT. ## Recorded results | Method | F1 Score | Result date | Report paper | | --- | --- | --- | --- | | WDR | 95.7 | Not recorded | 68fa6fe91e209131cea481c0 | | WDR | 94 | Not recorded | 68fa6fe91e209131cea481c0 | | WDR | 93.6 | Not recorded | 68fa6fe91e209131cea481c0 | | WDR | 93.1 | Not recorded | 68fa6fe91e209131cea481c0 | | WDR | 92.5 | Not recorded | 68fa6fe91e209131cea481c0 | | WDR | 91.1 | Not recorded | 68fa6fe91e209131cea481c0 | | FGWS | 90.6 | Not recorded | 68fa6fe91e209131cea481c0 | | FGWS | 89.5 | Not recorded | 68fa6fe91e209131cea481c0 | | FGWS | 89.5 | Not recorded | 68fa6fe91e209131cea481c0 | | FGWS | 88.9 | Not recorded | 68fa6fe91e209131cea481c0 | | FGWS | 88.7 | Not recorded | 68fa6fe91e209131cea481c0 | | FGWS | 87 | Not recorded | 68fa6fe91e209131cea481c0 | | WDR | 86.7 | Not recorded | 68fa6fe91e209131cea481c0 | | WDR | 86.4 | Not recorded | 68fa6fe91e209131cea481c0 | | WDR | 81 | Not recorded | 68fa6fe91e209131cea481c0 | | FGWS | 68.6 | Not recorded | 68fa6fe91e209131cea481c0 | | FGWS | 55.8 | Not recorded | 68fa6fe91e209131cea481c0 | | FGWS | 55.8 | Not recorded | 68fa6fe91e209131cea481c0 | ### WDR F1 Score: 95.7. Result ID: 699102153a6c31484904fb45. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"TextFooler","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### WDR F1 Score: 94. Result ID: 699102153a6c31484904fb3f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"LSTM","Attack":"PWWS","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### WDR F1 Score: 93.6. Result ID: 699102153a6c31484904fb3d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"PWWS","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### WDR F1 Score: 93.1. Result ID: 699102153a6c31484904fb67. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"PWWS"}. This record does not assert independent reproduction. ### WDR F1 Score: 92.5. Result ID: 699102153a6c31484904fb43. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"BERT","Attack":"PWWS","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### WDR F1 Score: 91.1. Result ID: 699102153a6c31484904fb41. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"CNN","Attack":"PWWS","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### FGWS F1 Score: 90.6. Result ID: 699102153a6c31484904fb42. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"CNN","Attack":"PWWS","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### FGWS F1 Score: 89.5. Result ID: 699102153a6c31484904fb3e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"PWWS","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### FGWS F1 Score: 89.5. Result ID: 699102153a6c31484904fb68. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"PWWS"}. This record does not assert independent reproduction. ### FGWS F1 Score: 88.9. Result ID: 699102153a6c31484904fb40. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LSTM","Attack":"PWWS","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### FGWS F1 Score: 88.7. Result ID: 699102153a6c31484904fb44. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"BERT","Attack":"PWWS","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### FGWS F1 Score: 87. Result ID: 699102153a6c31484904fb46. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"TextFooler","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### WDR F1 Score: 86.7. Result ID: 699102153a6c31484904fb49. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"IGA","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### WDR F1 Score: 86.4. Result ID: 699102153a6c31484904fb47. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"BAE","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### WDR F1 Score: 81. Result ID: 699102153a6c31484904fb69. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"BAE"}. This record does not assert independent reproduction. ### FGWS F1 Score: 68.6. Result ID: 699102153a6c31484904fb4a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"IGA","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### FGWS F1 Score: 55.8. Result ID: 699102153a6c31484904fb48. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"BAE","Detector Trained On":"DistilBERT, AG News, PWWS"}. This record does not assert independent reproduction. ### FGWS F1 Score: 55.8. Result ID: 699102153a6c31484904fb6a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"DistilBERT","Attack":"BAE"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.