REVIEW 17 cited by
On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
abstract
Large language model-based multi-agent systems have shown great abilities across various tasks due to the collaboration of expert agents, each focusing on a specific domain. However, the impact of clumsy or even malicious agents--those who frequently make errors in their tasks--on the overall performance of the system remains underexplored. This paper investigates: (1) What is the resilience of various system structures (e.g., A$\rightarrow$B$\rightarrow$C, A$\leftrightarrow$B$\leftrightarrow$C) under faulty agents, on different downstream tasks? (2) How can we increase system resilience to defend against these agents? To simulate faulty agents, we propose two approaches--AutoTransform and AutoInject--which introduce mistakes into the agents' responses. Experiments on four downstream tasks using six systems show that the "hierarchical" structure, i.e., A$\rightarrow$(B$\leftrightarrow$C), exhibits superior resilience with the lowest performance drop of 5.5%, compared to 10.5% and 23.7% of other two structures. To further improve resilience, we introduce (1) Challenger, that introduces a mechanism for each agent to challenge others' outputs, and (2) Inspector, an additional agent to review and correct messages, recovering up to 96.4% errors made by faulty agents. Our code and data are available at https://github.com/CUHK-ARISE/MAS-Resilience.
Forward citations
Cited by 17 Pith papers
-
Prompt Infection: LLM-to-LLM Prompt Injection within Multi-Agent Systems
Prompt injection attacks can self-replicate across LLM agents in multi-agent systems, enabling data theft, misinformation, and system disruption while propagating silently.
-
Hierarchical Attacks for Multi-Modal Multi-Agent Reasoning
HAM³ achieves up to 78.3% attack success rate on the GQA benchmark by hierarchically attacking perception, communication, and reasoning layers in multi-modal multi-agent systems.
-
Attacks and Mitigations for Distributed Governance of Agentic AI under Byzantine Adversaries
Identifies concrete attacks from a malicious Provider on SAGA and proposes SAGA-BFT, SAGA-MON, SAGA-AUD, and SAGA-HYB mitigations offering different security-performance trade-offs.
-
PropGuard: Safeguarding LLM-MAS via Propagation-Aware Exploration and Remediation
PropGuard is a propagation-aware framework for LLM-MAS that constructs dual-view spatio-temporal graphs, employs a GE-GRPO inspector to recover suspicious subgraphs, and applies source-guided remediation to lower atta...
-
Improving the Efficiency of Language Agent Teams with Adaptive Task Graphs
LATTE coordinates LLM agent teams with an evolving shared task graph, cutting token use, time, and failures while matching or beating accuracy of MetaGPT, leader-worker, and static methods.
-
Who&When Pro: Can LLMs Really Attribute Failures in AI Agents?
A warm-start error-injection pipeline yields 12,326 golden-labeled multimodal agent failures, and current LLMs remain weak at step-and-mode failure attribution.
-
Opinion Polarization in LLM-Based Social Networks: Manipulation and Mitigation
LLM-agent simulations show limited-budget adversaries can amplify polarization in social networks and that reactive and proactive mitigations do not fully restore baseline polarization.
-
StepFinder: A Temporal Semantic Framework for Failure Attribution in Multi-Agent Systems
StepFinder turns execution logs into temporal semantic sequences via LLMs then uses temporal modeling plus attention to attribute failures to specific steps more accurately and 79% faster than direct LLM methods on th...
-
Chain-of-Models: Cross-Model Auditing for Bias-Robust LLM Judges
Per-bias selection of a cross-family LLM auditor lifts biased-judgment accuracy from 0.805/0.824 baselines to 0.884.
-
Deterministic Inference across Tensor Parallel Sizes That Eliminates Training-Inference Mismatch
Tree-Based Invariant Kernels fix the floating-point reduction order across GPUs, making LLM logits and sampled tokens bitwise identical for tensor-parallel sizes 1/2/4/8 and exactly matching vLLM (TP=4) with FSDP (TP=1).
-
When Autonomy Goes Rogue: Preparing for Risks of Multi-Agent Collusion in Social Systems
In a 1,000-agent social simulation, decentralized groups of malicious AI agents spread more misinformation and commit more fraud than centralized groups, and they adapt to evade content moderation.
-
To trust or not to trust: Attention-based Trust Management for LLM Multi-Agent Systems
Introduces six-dimension trustworthiness definition and attention-based A-Trust score with a TMS to improve LLM-MAS robustness against malicious or unreliable messages.
-
Hallucination as Context Drift: Synchronization Protocols for Multi-Agent LLM Systems
Context drift between agents causes hallucinations in multi-agent LLMs; the Shared State Verification Protocol reduces them more effectively than full-broadcast synchronization with 58% fewer API calls.
-
OpenEvoShield: Dual Non-Stationary Continual Defense for Open-World Multi-Agent System Attacks
OpenEvoShield claims to defend LLM multi-agent systems against evolving and novel attacks by combining asymmetric-rate continual learning with energy-based OOD detection.
-
Emergent Social Intelligence Risks in Generative Multi-Agent Systems
Generative multi-agent systems exhibit emergent collusion and conformity behaviors that cannot be prevented by existing agent-level safeguards.
-
Before Humans Join the Team: Diagnosing Coordination Failures in Healthcare Robot Team Simulation
LLM-agent simulations of hierarchical healthcare robot teams show team structure as the primary bottleneck for coordination success, more than model capability or added context, while revealing a trade-off between age...
-
Agentic AI Security: Threats, Defenses, Evaluation, and Open Challenges
A survey that taxonomizes threats to agentic AI, reviews benchmarks and evaluation methods, discusses technical and governance defenses, and identifies open challenges.
Discussion (0). Sign in to comment.