The paper proposes the Explicit Belief Multi-Agent System (EB‑MAS) to address safety in large language model (LLM) powered multi‑agent systems. It treats belief as an engineering abstraction for evidence‑based commitments and externalizes belief states into a structured substrate, using belief operators to propose updates and a Cognitive Firewall to filter weakly evidenced changes. The authors also highlight risks such as manufactured consensus, collusion, and cascading misinformation, and discuss practical challenges in grounding, false positives, scalability, and ethical misuse.
arXiv:2606. 04990v2 Announce Type: replace-cross Abstract: Large language model (LLM)-based agents are evolving from passive text generators into autonomous systems capable of planning, tool use, retrieval, memory access, environmental interaction, and multi-agent collaboration.
arXiv:2606. 02282v1 Announce Type: new Abstract: Orchestrating Large Language Models into Multi-Agent Systems (LLM-MAS) has unlocked remarkable reasoning capabilities, yet emergent failures and hallucinations that resist characterisation block their deployment in safety-critical domains -- a gap made legally untenable by emerging AI regulation.
The paper discusses the trustworthiness of agentic AI systems built on large language models, highlighting new security and operational risks such as indirect prompt injection, memory contamination, and cross‑session data leakage. It categorizes failure modes, reviews mitigation strategies—including instruction hierarchies, context isolation, and constrained tool use—and introduces the Trustworthy Agent Development Lifecycle (TADL), a six‑phase framework for specification, design, training, evaluation, deployment, and monitoring. The authors note that TADL has not yet been empirically validated but offers a structured foundation for developing more secure and accountable agentic systems, and they call for improved benchmarks and future research priorities.
MAS-Guard is a unified framework designed to safeguard both single- and multi-agent large language model (LLM) systems. It introduces a local Agent Guardrail (A-Guard) layer that uses task-specific experts to evaluate agentic events, and a system-level Control Plane that applies deterministic structural checks and semantic assessments to routes, capabilities, and message content. The framework enforces interventions—allow, reflection, confirmation, and block—across benchmarks, reducing attack success rates by 71.5‑86.3% while maintaining utility on benign tasks.
arXiv:2608. 14590v1 Announce Type: new Abstract: LLM agents increasingly perform irreversible real-world actions, including database updates, API calls, file operations, and autonomous use of tools.
VEPA is a framework that enables the verified evolution of formal guardrails for lifelong large language model agents. It represents guardrails as formal policies of symbolic rules over typed agent events, and during deployment uses LLM‑guided feedback to propose policy patches that fix new failures while preserving correct behavior. A deterministic verifier accepts only structurally valid, conflict‑free policies, and an evidence‑aware authority layer activates verified rules only when sufficient supporting evidence exists, ensuring that uncertain updates do not directly affect guardrail decisions.
arXiv:2607. 08066v1 Announce Type: new Abstract: Chain-of-thought (CoT) monitoring is a promising safety mechanism for AI agents, based on the premise that visible reasoning traces can surface misaligned or deceptive behavior.
arXiv:2606. 10749v1 Announce Type: cross Abstract: Large language model (LLM) agents are rapidly moving from conversational interfaces to software components that plan, invoke tools, maintain memory, and act on external environments.
The paper introduces Preregistered Belief Revision Contracts (PBRC), a protocol that separates open communication from admissible epistemic change in deliberative multi-agent systems. PBRC fixes evidence triggers, revision operators, priority rules, and fallback policies, requiring that belief changes cite preregistered triggers and validated evidence tokens. The authors prove that PBRC prevents confidence inflation from conformity, preserves auditability, ensures epistemic accountability, and characterizes enforced belief trajectories under token-invariant contracts.
Persistent memory has enabled large language model (LLM) agents to store factual knowledge, prior decisions, reasoning histories, tool usage information, and context. While this has improved the agent's functionality and continuity across tasks, it has also introduced a new attack surface: the agent's own reasoning history.
arXiv:2606. 12797v1 Announce Type: new Abstract: Agentic large language model systems that autonomously invoke tools, maintain persistent memory, and execute multi-step plans are increasingly deployed in public-facing domains, including government services, healthcare triage, and financial advising.
arXiv:2608. 04289v1 Announce Type: new Abstract: Long-horizon agents increasingly use persistent memory and tools to take actions with external side effects.