3 papers
cs.LG2026
HyperFix: Combinatorial Nonlinear Correction for Task Vector Merging
Hyo Seo Kim, Ren Wang
Task vectors enable model merging without joint retraining. In practice, the subset of task vectors to be merged may vary, but many existing methods use scalar tuning for a particu…
cs.CR2026
Early Detection of Distributed Backdoors in Multi-Agent LLM Systems: A Characterization Study
Diego Fernandez Arias, Dev Prashant Mistry, Ren Wang +1
Multi-agent LLM systems can be attacked by a payload that no single agent ever holds in full: a poisoned tool hides encrypted fragments in its observations, spreads them across sev…
cs.CR2026
When Local Monitors Miss Compositional Harm: Diagnosing Distributed Backdoors in Multi-Agent Systems
Yibo Hu, Ren Wang
As multi-agent, tool-using LLM systems are deployed, a common safety net is a runtime monitor that checks each message, tool call, or step on its own. We show this net has a fundam…