Memory Contagion: Cross-Temporal Propagation of Evaluator Bias via Agent Memory

• URL: http://arxiv.org/abs/2606.23195v2
• Date: Wed, 24 Jun 2026 16:12:30 GMT
• Title: Memory Contagion: Cross-Temporal Propagation of Evaluator Bias via Agent Memory
• Abstract summary: Memory Contagion -- the cross-temporal propagation of evaluator bias through agent memory.<n>We show that when agents are trained or guided by biased evaluators, their experiences become biased.<n>Our findings expose a critical but contingent vulnerability in current agent memory designs.
• Score: 1.0152838128195467
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large Language Model (LLM) agents increasingly rely on memory systems to maintain long-term coherence. Recent work shows that agent memories degrade during continuous consolidation. However, existing research assumes memories are derived from unbiased experiences. In this work, we identify and formalize a novel phenomenon: Memory Contagion -- the cross-temporal propagation of evaluator bias through agent memory. We show that when agents are trained or guided by biased evaluators, their experiences become biased; when these trajectories are stored and consolidated into memory, the bias propagates to future agents retrieving from the same memory store, even when consolidation is perfect (oracle). Across two bias types (length preference, authority bias) and four experimental phases, we demonstrate: (1) Memory Contagion occurs for length bias even with perfect consolidation on older models (Gamma_A = 13.18, DeepSeek V4-Chat), while newer models (V4-Pro, Claude) are immune, proving both that biased input is a sufficient cause and that contagion is model-generation-dependent; (2) authority bias fails to propagate in all 15 controlled multi-seed experiments (Gamma_A = 0.00), revealing that not all evaluator biases can cross temporal boundaries through current memory architectures; (3) No observed safe threshold: length bias propagation is detected at contamination rates as low as p=0.2. Our findings expose a critical but contingent vulnerability in current agent memory designs and provide formal tools for measuring cross-temporal bias propagation.

Related papers

• Controlled Memory Interference in Continual LLM Agents [19.630594179909156]
  Long-term memory enables AI agents to maintain continuity across sessions, personalize behavior, and evolve through accumulated experience.<n>Existing systems mainly emphasize memory construction and relevance-based retrieval, but several memories may remain simultaneously relevant while differing in state, temporal validity, or authority.<n>We introduce Controlled Memory Interference (CMI), a controlled diagnostic and data-generation framework for studying how agent memory evolves under different memory relationships.
  arXiv  Detail & Related papers  (2026-08-07T09:46:19Z)
• MemOps: Benchmarking Lifecycle Memory Operations in Long-Horizon Conversations [50.24315431387575]
  We introduce MemOps, a benchmark that reformulates conversational memory as a sequence of lifecycle operations.<n>MemOps represents each memory event with a structured trace specifying its trigger, target, scope, state transition, and supporting evidence.<n>Across long-context, retrieval-based, parametric and managed-memory systems, MemOps disentangles failure modes that final-answer accuracy alone conceals.
  arXiv  Detail & Related papers  (2026-07-14T15:33:44Z)
• SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents [38.778004697710855]
  We introduce SubtleMemory, a benchmark for fine-grained relational memory discrimination in long-running AI agents.<n>We evaluate six standalone memory systems, two Claw-style agents with native memory modules, and three Claw-style agents with plugin memory modules.
  arXiv  Detail & Related papers  (2026-06-04T06:43:11Z)
• WorldMemArena: Evaluating Multimodal Agent Memory Through Action-World Interaction [72.1620416874118]
  Multimodal large language models are increasingly deployed as long-horizon agents.<n>Existing benchmarks measure recall over static dialogue, collapse memory into a single end-of-task accuracy, and reduce visual observations to captions.<n>We formulate multimodal agent memory as an Action-World Interaction Loop with an observable four-stage lifecycle.
  arXiv  Detail & Related papers  (2026-05-28T04:27:20Z)
• The Trap of Trajectory: Towards Understanding and Mitigating Spurious Correlations in Agentic Memory [7.707732051915869]
  Diagnosing agentic memory systems reveals that memory improves reasoning on clean inputs but amplifies reliance on spurious patterns when they are present.<n>We propose CAMEL, a plug-and-play calibration method that operates across diverse memory architectures at both write and retrieval time.<n>Overall, CAMEL offers a principled and lightweight solution toward more reliable agentic memory deployment.
  arXiv  Detail & Related papers  (2026-05-10T05:04:13Z)
• Belief Memory: Agent Memory Under Partial Observability [56.41506249481312]
  We propose BeliefMem, which shifts the memory paradigm from committing to a single conclusion per observation to retaining multiple candidate conclusions with their probabilities.<n> BeliefMem preserves the uncertainty that the deterministic paradigm discards, enabling the agent to act with high confidence.<n> Empirical evaluations on LoCoMo and ALFWorld benchmarks show that, even with limited data, BeliefMem achieves the best average performance.
  arXiv  Detail & Related papers  (2026-05-07T02:03:13Z)
• MemEvoBench: Benchmarking Memory MisEvolution in LLM Agents [78.95081012334116]
  Large Language Models (LLMs) with persistent memory enhances interaction continuity and personalization but introduces new safety risks.<n> contaminated or biased memory accumulation can trigger abnormal agent behaviors.<n>MemeEvoBench is the first benchmark evaluating memory safety in LLM agents against adversarial memory injection, noisy tool outputs, and biased feedback.
  arXiv  Detail & Related papers  (2026-04-17T07:29:52Z)
• MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Tasks [55.145729491377374]
  Existing evaluations of agents with memory typically assess memorization and action in isolation.<n>We introduce MemoryArena, a unified evaluation gym for benchmarking agent memory in multi-session Memory-Agent-Environment loops.<n> MemoryArena supports evaluation across web navigation, preference-constrained planning, progressive information search, and sequential formal reasoning.
  arXiv  Detail & Related papers  (2026-02-18T09:49:14Z)
• REMem: Reasoning with Episodic Memory in Language Agent [32.63834745610879]
  We present REMem, a framework for constructing and reasoning with episodic memory.<n>We show that REMem substantially outperforms state-of-the-temporal-art memory systems such as Mem0 and HippoRAG 2.<n>REMem also demonstrates more robust refusal behavior for unanswerable questions.
  arXiv  Detail & Related papers  (2026-02-13T23:54:55Z)
• MemoryGraft: Persistent Compromise of LLM Agents via Poisoned Experience Retrieval [5.734678752740074]
  MemoryGraft is a novel indirect injection attack that compromises agent behavior not through immediate jailbreaks, but by implanting malicious successful experiences into the agent's long-term memory.<n>We demonstrate that an attacker who can supply benign ingestion-level artifacts that the agent reads during execution can induce it to construct a poisoned RAG store.<n>When the agent later encounters semantically similar tasks, union retrieval over lexical templates and embedding similarity reliably surfaces these grafted memories, and the agent adopts the embedded unsafe patterns, leading to persistent behavioral drift across sessions.
  arXiv  Detail & Related papers  (2025-12-18T08:34:40Z)
• Memory in the Age of AI Agents [217.9368190980982]
  This work aims to provide an up-to-date landscape of current agent memory research.<n>We identify three dominant realizations of agent memory, namely token-level, parametric, and latent memory.<n>To support practical development, we compile a comprehensive summary of memory benchmarks and open-source frameworks.
  arXiv  Detail & Related papers  (2025-12-15T17:22:34Z)
• Agentic Learner with Grow-and-Refine Multimodal Semantic Memory [50.81667005063605]
  ViLoMem is a dual-stream memory framework that constructs compact, schema-based memory.<n>It encodes visual distraction patterns and logical reasoning errors, enabling MLLMs to learn from their successful and failed experiences.
  arXiv  Detail & Related papers  (2025-11-26T18:55:08Z)
• BURN: Backdoor Unlearning via Adversarial Boundary Analysis [73.14147934175604]
  Backdoor unlearning aims to remove backdoor-related information while preserving the model's original functionality.<n>We propose Backdoor Unlearning via adversaRial bouNdary analysis (BURN), a novel defense framework that integrates false correlation decoupling, progressive data refinement, and model purification.
  arXiv  Detail & Related papers  (2025-07-14T17:13:06Z)
• Collaborative Value Function Estimation Under Model Mismatch: A Federated Temporal Difference Analysis [55.13545823385091]
  Federated reinforcement learning (FedRL) enables collaborative learning while preserving data privacy by preventing direct data exchange between agents.<n>In real-world applications, each agent may experience slightly different transition dynamics, leading to inherent model mismatches.<n>We show that even moderate levels of information sharing significantly mitigate environment-specific errors.
  arXiv  Detail & Related papers  (2025-03-21T18:06:28Z)

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.