← back to paper

Review history

arxiv: 2608.00718 · 2 revisions

Adversarial Attacks in Multi-Agent LLM Pipelines: Unveiling Structural Vulnerabilities in Agentic AI Architectures

1. 2026-08-15 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 5.0
   139145 ms 238 in 12710 out 2026-08-15T15:17:52.267006+00:00
2. 2026-08-05 REJECT MODERATE v1.3.0-alltime-deepseek novelty 5.0
   184538 ms 9710 in 17638 out 2026-08-05T00:28:18.606776+00:00