Newest papers

After the Attacker Leaves: Benchmarking Carryover Attacks on LLM Agents

Zhiyue Xu, Xukai Wang, Zhiyuan Huang, Mingjie Zhan, Jian Zhao, Xu-Yao Zhang · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.28425 · cited by 0 works (OpenAlex) · open-access PDF

Korean

현대 LLM 에이전트는 두 채널을 통해 세션에 걸쳐 상태를 유지합니다 : 오래된 메모리 파일은 시스템 프롬프트에 조용히 전달되며 일정한 작업은 사용자 입력없이 에이전트를 깨우고 있습니다. 한 번의 간접 프롬프트 주입은 공격자가 떠난 후 세션을 통해 다시 화재 할 수 있습니다. 위협은 전송 공격을 말합니다. 기존의 프롬프트 주입 벤치마크는 악성 콘텐츠가 처음 관찰되는 세션에서만 공격을 평가하며 게이트가 언제 그리고 얼마나 자주 정착 된 유용 부하가 나중에 측정되지 않은 상태로 다시 화재 할 수 있습니다. CarryoverBench: 12 개의 개인 지원 시나리오와 3 개의 트리거 채널에 걸친 101 개의 공격 작업: 즉

Machine translation (Meta M2M100); the English original follows.

Abstract

Modern LLM agents persist state across sessions via two channels: long-lived memory files silently prepended to the system prompt, and scheduled tasks waking the agent without user input. A one-time indirect prompt injection committed to either channel can re-fire across sessions long after the attacker leaves, threats we term carryover attacks. Existing prompt-injection benchmarks evaluate attacks only in the session where malicious content is first observed, leaving the gate governing when and how often a planted payload later re-fires unmeasured. We present CarryoverBench: 12 personal-assistant scenarios and 101 attack tasks across three trigger channels: Instant, same-turn IPI; Autonomous, heartbeat-scheduled; and Latent, memory-resurfaced. A rule-based verifier scores planting and triggering as independent stages. Across five frontier models, planted payloads re-fire without measurable decay over five rounds; conditional on successful planting, the trigger stage often provides little additional protection. Porting the Latent channel to a second runtime, Hermes, attenuates but does not eliminate the effect, suggesting that carryover defense requires runtime-layer controls in addition to prompt-level interventions.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Session (web analytics), Computer science, Computer security, Payload (computing), Porting, State (computer science)

Is this paper real? This page was built from OpenAlex and arXiv records, not generated. Software can verify any reference in one call: GET https://store.agentexchange.work/gw/papers/verify?title=...&year=... (0.003 USD), fetch this record as JSON with /gw/papers/paper?id=2610.28425, or list what it cites with /gw/papers/references?id=.... Pay per call in USDC (x402) or with a 19 USD card key for 1,000 calls. Catalogue · llms.txt

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange