This paper adapts attacks and defenses from multi-agent systems literature to multi-agent collaborative filtering systems, evaluating their impact under varying connectivity.
Adapted attacks and defenses from multi-agent systems to multi-agent collaborative filtering and studied their impact under varying connectivity.
Before reading this…
Applications
To understand this paper, make sure you know these concepts first:
Multi-agent collaborative filtering (CF) systems coordinate autonomous LLM-powered user and item agents through natural-language interaction to refine preferences and generate recommendations. These systems inherit vulnerabilities from both their data-driven nature and their multi-agent interactions, which manifest in distinct ways. Understanding how connectivity modulates vulnerability in these systems could facilitate the development of more robust recommendation pipelines. In this work, we adapt attacks and defenses from the general multi-agent systems (MAS) literature to the agent-based CF setting, evaluating them under systematically varied connectivity in the AgentCF framework, where CF connectivity is characterized along two axes: (i) candidate count (the number of item candidates per turn per user, measuring user-side interaction density) and (ii) catalog concentration (the degree of item catalog overlap across users). Our contributions include: (1) Adaptation: we reproduce MAS-inspired attacks and defenses in the agentic CF domain, confirming partial transferability of original observations. (2) Characterization: we characterize how the two aspects of connectivity shape attack and defense outcomes, revealing role asymmetries between user and item agents, non-monotonic temporal dynamics in attack efficacy, and divergent patterns across dissemination and extraction attack goals. Additionally, as an exploratory extension, we assess the applicability of epidemic-inspired static metrics in ranking CF configurations by expected attack outcome, potentially enabling cost-efficient robustness assessment. Implementation is available at https://github.com/anjunhu/ConnACF
Autonomous Repair for Multi-Agent Systems via Monte-Carlo Tree Search
Proposed MARS, a search-based framework for multi-agent system repair using Mont…
SafeFlow: Semantic Information-Flow Control for Blocking Malicious Propagation in Multi-Agent System…
This paper introduces SafeFlow, a defense framework for multi-agent systems that…
Robust Estimation of Sparse Numerical Vectors under Local Differential Privacy
This paper proposes Randomized Projection with Clipping (RPC) to enhance robustn…
MARS: Multi-Agent Re-ranking for Repeat-Order Food Delivery Recommendation
This paper introduces MARS, a framework for repeat-order food delivery recommend…
A Framework of User Experience Principles for Human-AI Agent Interaction in the Workplace
This paper identifies and validates eight core UX principles for human-AI agent…
MR-TGN: A Meta-Role Temporal Graph Network for Team-Level Intent Prediction in Multi-Agent Systems
This paper proposes a Meta-Role Temporal Graph Network (MR-TGN) framework for co…
Personalized Recommendation Tool Learning via Autonomous Language Agents
A new framework, PRTA, is proposed for full-ranking recommendation tasks using l…
Beyond Component Testing: Validating Agentic AI Systems
This paper surveys 257 papers on validating agentic AI systems, identifying matu…