Paper recorded by Signals 4 on 2026-10-30 in cs.AI. Abstract reproduced from arXiv; link to the original below.
Published 2026-10-30 on arXiv · recorded by Signals 4 on 2026-11-02
Category: cs.AI · 人工智能 · first seen 2026-11-02
Recognized as a top threat to AI agents, prompt injection allows instructions from untrusted data (email, documents, websites) to manipulate the agent's operation. The existing best-defended LLM, Meta-SecAlign, is trained with the DPO loss but still suffers from attack success rates (ASRs) approaching 100% under adaptive prompt injections. One reason is that DPO anchors the model around the undefe