ResearchBenchmarksAdversarial Robustness on ActorAttack (out-of-domain)Adversarial Robustness on ActorAttack (out-of-domain)Follow0.435ASRBaseline0.11260.19630.280.3637May 22, 2025Evaluation resultsMethodMethodLinksASRVarianceBaselineTarget Model=zephyr-7b...Target Model=zephyr-7b-beta2025.050.4350MTSA-T1Target Model=zephyr-7b...Target Model=zephyr-7b-beta, Iteration=12025.050.245-0.4367MTSA-T2Target Model=zephyr-7b...Target Model=zephyr-7b-beta, Iteration=22025.050.17-0.6009MTSA-T3Target Model=zephyr-7b...Target Model=zephyr-7b-beta, Iteration=32025.050.125-0.7126