Pith. sign in

← back to paper

Review history

arxiv: 2606.02240 · 2 revisions

AgentRedBench: Dynamic Redteaming and Integration-Aware Defense for LLM Agents over SaaS Integrations

  1. 2026-08-02 REJECT HIGH v1.3.0-alltime-deepseek novelty 6.0
    165931 ms 20317 in 18159 out 2026-08-02T12:32:21.970458+00:00
  2. 2026-06-28 UNVERDICTED UNKNOWN v0.9.1-grok novelty 7.0
    26513 ms 5852 in 1260 out 2026-06-28T13:52:33.511336+00:00