# White-box robustness against single point failure attacks on WikiText-2 Source: [White-box robustness against single point failure attacks on WikiText-2]() Collection index: [llms.txt]() Data updated: 2026-03-18T08:39:30.103Z Evaluation of model robustness on the WikiText-2 dataset against white-box single point failure attacks, reporting original and post-attack perplexity (PPL). 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 | White-box robustness against single point failure attacks | | Dataset (including any stated split) | WikiText-2 | | Metric | Original Perplexity (PPL) | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 8 of 8 | | 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 - [Original Perplexity (PPL)](): Lower is better; unit: SCALAR. - [Attack Bits](): Higher is better; unit: SIZE. - [Post-Attack Perplexity (PPL)](): Lower is better; unit: SCALAR. ## Recorded results | Method | Original Perplexity (PPL) | Result date | Report paper | | --- | --- | --- | --- | | Baseline | 5.03 | Not recorded | 69ba31870636e5bf4afc07b5 | | RADAR | 5.03 | Not recorded | 69ba31870636e5bf4afc07b5 | | RoR | 5.03 | Not recorded | 69ba31870636e5bf4afc07b5 | | Baseline | 5.41 | Not recorded | 69ba31870636e5bf4afc07b5 | | RADAR | 5.41 | Not recorded | 69ba31870636e5bf4afc07b5 | | RoR | 5.41 | Not recorded | 69ba31870636e5bf4afc07b5 | | FaR | 6.56 | Not recorded | 69ba31870636e5bf4afc07b5 | | FaR | 8 | Not recorded | 69ba31870636e5bf4afc07b5 | ### Baseline Original Perplexity (PPL): 5.03. Result ID: 69ba64a447f34ce924a457a6. [Reporting paper]() Recorded evaluation configuration: {"Model":"Llama-2-7B"}. This record does not assert independent reproduction. ### RADAR Original Perplexity (PPL): 5.03. Result ID: 69ba64a447f34ce924a457a8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-7B"}. This record does not assert independent reproduction. ### RoR Original Perplexity (PPL): 5.03. Result ID: 69ba64a447f34ce924a457a9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Llama-2-7B"}. This record does not assert independent reproduction. ### Baseline Original Perplexity (PPL): 5.41. Result ID: 69ba64a447f34ce924a457aa. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### RADAR Original Perplexity (PPL): 5.41. Result ID: 69ba64a447f34ce924a457ac. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### RoR Original Perplexity (PPL): 5.41. Result ID: 69ba64a447f34ce924a457ad. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### FaR Original Perplexity (PPL): 6.56. Result ID: 69ba64a447f34ce924a457ab. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-7B"}. This record does not assert independent reproduction. ### FaR Original Perplexity (PPL): 8. Result ID: 69ba64a447f34ce924a457a7. [Reporting paper]() Recorded evaluation configuration: {"Model":"Llama-2-7B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.