MemMorph introduces a novel memory poisoning attack that biases LLM agent tool selection by injecting crafted records into the agent's long-term memory, achieving high success rates even against moder…

Evo-Attacker introduces a memory-augmented reinforcement learning framework to perform generalized, long-horizon tool attacks on LLM-MAS, significantly outperforming existing methods.

The paper proposes a unified framework to evaluate how different types of memory transfer benefit multi-trajectory inference for tool-use LLM agents, finding that the optimal memory method depends cri…

The paper introduces and evaluates 'sleeper memory poisoning,' a delayed adversarial attack that corrupts an LLM agent's persistent memory by manipulating external context, demonstrating that these po…

The paper introduces MemSecBench, a benchmark for evaluating the lifecycle security of agent memory systems against malicious semantics, and reports results across various configurations.

This paper systematically studies memory poisoning attacks in LLM agents, identifying multiple vulnerabilities and proposing a new benchmark to assess the risk.

This paper identifies a behavioral invariant in LLM agents under memory poisoning attacks, which can be used to detect such attacks with high accuracy.

This paper introduces AutoMem, a framework that automates memory management as a trainable skill for large language models, improving performance up to 2x-4x on long-horizon tasks.

The paper proposes SAGE, a novelty-aware gate that efficiently controls memory updates in agentic LLMs by classifying new facts as clearly novel, clearly redundant, or uncertain, thereby significantly…

The paper introduces CyBiasBench, a comprehensive benchmark that quantifies the inherent, agent-specific bias in LLM agents' attack selection patterns in cybersecurity scenarios.

The paper proposes MemPoison, a novel memory poisoning attack that injects triggerable backdoors into LLM agents' long-term memory through dialogue interactions, achieving high success rates by bypass…

The paper introduces MemPoison, a novel memory poisoning attack that successfully injects triggerable backdoors into LLM agents' long-term memory through conversational interactions, achieving high at…

Proposed MemSyco-Bench benchmark for evaluating memory-induced sycophancy in agent systems, measuring when and how valid memories should be used.

The paper introduces MemCog, a Memory-as-Cognition system that integrates memory access directly into the reasoning process, significantly improving agent performance, especially in proactive memory r…

The paper introduces a new security benchmark and framework to defend LLM agents against 'cognitive poisoning,' where malicious tools build trust through benign feedback before executing a harmful fin…

The paper identifies the Misattribution Gap, showing that memory-layer attacks (Semantic Norm Drift) can mimic model failure in multi-agent AI systems, and proposes novel detection and mitigation tech…

Defense training for LLM agents, intended to improve safety, systematically degrades their core competence, leading to unreliability in multi-step tasks.

This paper analyzes tool-calling in LLM agents, demonstrating that evaluation results are highly sensitive to implementation details and proposing new techniques to significantly improve the efficienc…

The paper introduces Trojan Hippo, a persistent memory attack that exfiltrates sensitive data from LLM agents by planting dormant payloads into long-term memory, and develops a comprehensive framework…