Showing cs.CRShow all
2 papers · 1 filter
cs.CR2026
SilentCall: Hidden Tool-Call Backdoors in Open-Weight Agents, and How to Catch Them
Bhanu Pallakonda, Mikkel Hindsbo, Sina Ehsani +1
Open-weight tool-calling agents are adopted on evidence of merit, usually benchmark scores and a record of reliable use. We show that a model publisher can train an agent that earn…
cs.CR2026
Sleeper Cell: Injecting Latent Malice Temporal Backdoors into Tool-Using LLMs
Bhanu Pallakonda, Mikkel Hindsbo, Sina Ehsani +1
The proliferation of open-weight Large Language Models (LLMs) has democratized agentic AI, yet fine-tuned weights are frequently shared and adopted with limited scrutiny beyond lea…