Self-evolving LLM agents can be attacked by planting individually benign experiences that jointly weaken their safety boundaries.
Findings of the Association for Computational Linguistics: ACL 2026 , pages=
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Benign Alone, Harmful Together: Exploiting Experience Composition in Self-Evolving LLM Agents
Self-evolving LLM agents can be attacked by planting individually benign experiences that jointly weaken their safety boundaries.