# Jailbreak and Phishing Prompt Classification on 60 Malicious Queries Source: [Jailbreak and Phishing Prompt Classification on 60 Malicious Queries]() Collection index: [llms.txt]() Data updated: 2026-04-02T09:17:19.036Z Evaluation of model performance on classifying prompts as either jailbreak or phishing attempts using the 60 Malicious Queries dataset, measured by Accuracy. 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 | Jailbreak and Phishing Prompt Classification | | Dataset (including any stated split) | 60 Malicious Queries | | Metric | Accuracy (Jailbreak/Phishing) | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 1 of 1 | | 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 - [Accuracy (Jailbreak/Phishing)](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy (Jailbreak/Phishing) | Result date | Report paper | | --- | --- | --- | --- | | Xu et al. | 92 | Not recorded | 68fa7a081e209131cea80a81 | ### Xu et al. Accuracy (Jailbreak/Phishing): 92. Result ID: 69ce341b47f34ce924b79c65. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Attack Vectors":"Jailbreaking but not specific to phishing","Models":"Vicuna, Llama, GPT-3.5","Best Model":"Bergeron the immune system and RoBERTa Classifier"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.