← back to paper

Review history

arxiv: 2506.15253 · 2 revisions

RAS-Eval: A Comprehensive Benchmark for Security Evaluation of LLM Agents in Real-World Environments

1. 2026-08-15 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
   102197 ms 3309 in 4120 out 2026-08-15T19:38:33.316327+00:00
2. 2026-08-07 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
   138543 ms 11136 in 14326 out 2026-08-07T00:00:34.783463+00:00