Skip to content
THE AI WIREINTELLIGENCE THAT MATTERS
Agents

ASLEval: Measuring Privacy Exposure Displacement in LLM Agent Sessions

arXiv:2609.18864v1 Announce Type: cross Abstract: Privacy evaluations of tool-using LLM agents often inspect a designated action, final response, or attacker report. These local proxies can miss unauthorized exposure elsewhere in a multi-step session and lack

arXiv cs.AI··Updated just now·34 sightings