Research index
A focused index for recent papers, topic groupings, concepts, and trend signals related to multi-agent security.
Recent publications
BazaarBench is a novel simulated decentralized consumer-to-consumer (C2C) marketplace benchmark designed to evaluate safety and delegation failures of Large Language Model (LLM) agents acting autonomously in buying and selling scenarios.
Open article pageThe paper addresses the trade-off in language model compliance known as the Pushback Paradox: models that always comply are exploitable, whereas models that resist cannot be stopped, impacting their controllability.
Open article pageThe paper addresses the challenge of agentic abstention in large language model (LLM) agents, specifically their difficulty recognizing when tasks are infeasible and should be declined.
Open article pageResearch trends
3772 papers
2544 papers
1162 papers