Understanding and Mitigating Inference-Time Overreliance Using Agentic Memory
cs.AI
Submitted: 2026-10-05
Updated: 2026-10-05
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
- Micro-Act: Mitigating Knowledge Conflict in LLM-based RAG via Actionable Self-Reasoning
- Flying Pigs, FaR and Beyond: Evaluating LLM Reasoning in Counterfactual Worlds
- Data to Defense: The Role of Curation in Customizing LLMs Against Jailbreaking Attacks
- ShadowMerge: A Novel Poisoning Attack on Graph-Based Agent Memory via Relation-Channel Conflicts
- Uncovering Vulnerabilities of LLM-Assisted Cyber Threat Intelligence
- Benchmarking LLM-Assisted Blue Teaming via Standardized Threat Hunting
- EquiMem: Calibrating Shared Memory in Multi-Agent Debate via Game-Theoretic Equilibrium
- MemGPT: Towards LLMs as Operating Systems
- ER-MIA: Black-Box Adversarial Memory Injection Attacks on Long-Term Memory-Augmented Large Language Models
- PersistBench: When Should Long-Term Memories Be Forgotten by LLMs?
- Zep: A Temporal Knowledge Graph Architecture for Agent Memory
- POLAR: Automating Cyber Threat Prioritization through LLM-Powered Assessment
- The Value of Variance: Mitigating Debate Collapse in Multi-Agent Systems via Uncertainty-Driven Policy Optimization
- The Trap of Trajectory: Towards Understanding and Mitigating Spurious Correlations in Agentic Memory
- OEP: Poisoning Self-Evolving LLM Agents via Locally Correct but Non-Transferable Experiences
- Agent Workflow Memory
- A-MemGuard: A Proactive Defense Framework for LLM-Based Agent Memory
- Certifiably Robust RAG against Retrieval Corruption
- Useful Memories Become Faulty When Continuously Updated by LLMs
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems