Before Agents Speak: Pre-hoc Failure Risk Inference in Multi-Agent Systems

• URL: http://arxiv.org/abs/2607.26836v1
• Date: Wed, 29 Jul 2026 12:26:41 GMT
• Title: Before Agents Speak: Pre-hoc Failure Risk Inference in Multi-Agent Systems
• Abstract summary: LLM-based multi-agent systems (MAS) have exhibited remarkable capabilities in collaborative reasoning and decision-making.<n> localized hallucinations can propagate along agent communication chain, amplify through interactions, and ultimately trigger cascading failures.<n>Existing countermeasures predominantly follow a post-hoc paradigm, identifying failures only after unsafe behaviors emerge.<n>HalluProp estimates individual agent failures and emergent system-level hallucination risks before inter-agent interaction.
• Score: 16.529918071209643
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: LLM-based multi-agent systems (MAS) have exhibited remarkable capabilities in collaborative reasoning and decision-making, yet their interconnected communications introduce new systemic risk: localized hallucinations can propagate along agent communication chain, amplify through interactions, and ultimately trigger cascading failures. Existing countermeasures predominantly follow a post-hoc paradigm, identifying failures only after unsafe behaviors emerge, by which time harmful effects may have already spread throughout the agent network. To tackle this problem, we investigate a complementary pre-hoc approach and propose HalluProp, a Propagation-aware Hallucination inference framework that estimates individual agent failures and emergent system-level hallucination risks before inter-agent interaction. First, we model intrinsic hallucination risks by identifying fine-grained semantic misalignment between agent roles and task queries. We then characterize inter-agent risk propagation by modeling both semantic influence and communication topology. Finally, we integrate these two risks via a differentiable Noisy-OR inference mechanism to derive a systemic diagnosis. Extensive experiments show that HalluProp accurately localizes faulty agents, achieving an average AUROC of 84.6%, while enabling sub-second diagnosis with over $65\times$ speedup over post-hoc methods. By facilitating early intervention through upstream screening, HalluProp effectively complements post-hoc methods, highlighting the potential of pre-hoc risk inference for building more reliable multi-agent systems.

Related papers

• Poisoning Agentic Alpha: Adversarial Vulnerabilities Across Roles and Architectures in Multi-Agent Trading Systems [103.90831279037094]
  LLM-based multi-agent trading systems are moving rapidly from research prototypes to live deployments that control real assets.<n>A corrupted signal can propagate to the final decision and translate into realized financial loss.<n>We present the first systematic empirical study in the financial domain to characterize how an adversarial signal enters a multi-agent trading system.
  arXiv  Detail & Related papers  (2026-08-25T05:07:37Z)
• Collective Hallucination in Multi-Agent LLMs:Modeling and Defense [0.5076419064097734]
  Hallucinations in large language models (LLMs) create heightened risks in multi-agent settings.<n>This paper models hallucination as a system-level, time-evolving process across a network of interacting LLM agents.<n>We introduce an interaction-aware control method that combines confidence-weighted aggregation, adaptive impact regulation, external claim verification, and selective isolation of unreliable agents.
  arXiv  Detail & Related papers  (2026-06-06T02:04:03Z)
• Multi-Agent Systems are Mixtures of Experts: Who Becomes an Influencer? [31.047683581225087]
  We show that Friedkin-Johnsen (FJ) parameters are input-dependent, turning multi-agent deliberation into a mixture of experts.<n>We analyze how influence is established through observable proxies: agents' self-assessed confidence, their perceived confidence, and initial alignment with other agents' views.
  arXiv  Detail & Related papers  (2026-05-25T15:10:48Z)
• Catching the Infection Before It Spreads: Foresight-Guided Defense in Multi-Agent Systems [8.60940551895875]
  Multi-Agent Systems (MASs) enable collaborative complex problem solving through specialized agents.<n>MASs are vulnerable to infectious jailbreak, where compromising a single agent can spread to others, leading to widespread compromise.<n>Existing defenses counter this by training a more contagious cure factor, biasing agents to retrieve it over virus adversarial examples.<n>We propose Foresight-Guided Local Purification (FLP) framework, where each agent reasons over future interactions to track behavioral evolution and eliminate infections.
  arXiv  Detail & Related papers  (2026-05-03T07:38:42Z)
• Taming Actor-Observer Asymmetry in Agents via Dialectical Alignment [59.536125286960186]
  Multi-agent frameworks assigning specialized roles are increasingly adopted to enable self-reflection and mutual auditing.<n>We find it simultaneously induces a human-like cognitive bias known as Actor-Observer Asymmetry.<n>ReTAS guides agents to synthesize conflicting viewpoints into an objective consensus.
  arXiv  Detail & Related papers  (2026-04-21T15:05:58Z)
• CIA: Inferring the Communication Topology from LLM-based Multi-Agent Systems [52.48183625867145]
  Multi-Agent Systems (MAS) have demonstrated remarkable capabilities in solving complex tasks.<n>Central to MAS is the communication topology which governs how agents exchange information internally.<n>In this paper, we investigate a critical privacy risk: MAS communication topologies can be inferred under a restrictive black-box setting.
  arXiv  Detail & Related papers  (2026-04-14T08:48:15Z)
• Auditing Cascading Risks in Multi-Agent Systems via Semantic-Geometric Co-evolution [5.051136155296043]
  We propose a principled framework for cascading-risk detection grounded in semantic--geometric co-evolution.<n>By coupling semantic flow signals with graph geometry, the framework learns the normal co-evolutionary dynamics of trusted collaboration.
  arXiv  Detail & Related papers  (2026-03-04T16:45:26Z)
• AgentDoG: A Diagnostic Guardrail Framework for AI Agent Safety and Security [126.49733412191416]
  Current guardrail models lack agentic risk awareness and transparency in risk diagnosis.<n>We propose a unified three-dimensional taxonomy that categorizes agentic risks by their source (where), failure mode (how), and consequence (what)<n>We introduce a new fine-grained agentic safety benchmark (ATBench) and a Diagnostic Guardrail framework for agent safety and security (AgentDoG)
  arXiv  Detail & Related papers  (2026-01-26T13:45:41Z)
• Agentic Uncertainty Quantification [76.94013626702183]
  We propose a unified Dual-Process Agentic UQ (AUQ) framework that transforms verbalized uncertainty into active, bi-directional control signals.<n>Our architecture comprises two complementary mechanisms: System 1 (Uncertainty-Aware Memory, UAM), which implicitly propagates verbalized confidence and semantic explanations to prevent blind decision-making; and System 2 (Uncertainty-Aware Reflection, UAR), which utilizes these explanations as rational cues to trigger targeted inference-time resolution only when necessary.
  arXiv  Detail & Related papers  (2026-01-22T07:16:26Z)
• Can an Individual Manipulate the Collective Decisions of Multi-Agents? [53.01767232004823]
  M-Spoiler is a framework that simulates agent interactions within a multi-agent system to generate adversarial samples.<n>M-Spoiler introduces a stubborn agent that actively aids in optimizing adversarial samples.<n>Our findings confirm the risks posed by the knowledge of an individual agent in multi-agent systems.
  arXiv  Detail & Related papers  (2025-09-20T01:54:20Z)
• HiddenBench: Assessing Collective Reasoning in Multi-Agent LLMs via Hidden Profile Tasks [12.203366267017737]
  We introduce HiddenBench, the first benchmark for evaluating collective reasoning in multi-agent LLMs.<n>To ground the benchmark, we formalize the paradigm with custom tasks and show that GPT-4.1 groups fail to integrate distributed knowledge.<n>We then construct the full benchmark, spanning 65 tasks drawn from custom designs, prior human studies, and automatic generation.
  arXiv  Detail & Related papers  (2025-05-15T19:22:54Z)
• Preventing Rogue Agents Improves Multi-Agent Collaboration [21.955058255432974]
  We propose to monitor agents during action prediction and intervene when a future error is likely to occur.<n>Experiments on WhoDunitEnv, code generation tasks and the GovSim environment for resource sustainability show that our approach leads to substantial performance gains.
  arXiv  Detail & Related papers  (2025-02-09T18:35:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.