arXiv AI
Jun 17
MemTrace: Probing What Final Accuracy Misses in Long-Term Memory
arXiv:2606. 17328v1 Announce Type: new Abstract: LLM agents increasingly maintain long-term memory of user facts across sessions.
When Retrieval Is Not Enough: Obsolete Facts in Agent Memory
The study shows that in memory‑agent benchmarks, simply retrieving the correct fact does not guarantee accurate answers. On the FactConsolidation task, increasing the evidence budget from 64 to 512 tokens caused accuracy to drop, even though the newest statement was present in every retrieved set. Experiments with Qwen models revealed that replacing an unrelated statement with an obsolete competing statement drastically reduced accuracy, indicating that outdated information can be more harmful than irrelevant evidence.
Related stories
arXiv AI
Aug 10
TEPA: Revoking Stale Memories for Conflict-Robust Language Agents
arXiv:2608. 07429v1 Announce Type: new Abstract: Long-term memory enables language agents to reuse past facts, preferences, and task experience.
arXiv Machine Learning
Jul 20
Presentation, Not Mechanism: A Render Confound in Deprecation-Aware Memory Evaluation
arXiv:2607. 16019v1 Announce Type: new Abstract: AI systems increasingly retrieve from records that revise themselves: issue threads, encyclopedic histories, policy logs, and long conversations.
arXiv AI
Oct 15
Memory Echo in Single-Agent Long-Term Memory: Lifecycle Evaluation and Provenance-Guided Mitigation
arXiv:2610.17763v1 Announce Type: new Abstract: Long-term memory supports agents across sessions. Repeated retrieval and write-back can let one historical claim occupy multiple retrieval positions an...
arXiv AI
Sep 30
Memory Is a Derivation: The Distributed-Evidence Paradox in Long-Term Agents
The paper introduces the Distributed‑Evidence Paradox, where long‑running LLM agents compress past interactions into persistent memories that may not be fully supported by the interaction history. It defines three key requirements—evidence scope, compositional validity, and admission reliability—and proposes DerivAudit, a framework that checks whether a memory is truly supported by the available history. Experiments on two memory corpora show that expanding the evidence base can recover support for many memories, yet many remain unsupported, and broader evidence alone does not guarantee reliable admission.
arXiv AI
Sep 10
What Eviction Destroys: A Restore-Counterfactual Audit of Forgetting in Agent Memory
arXiv:2609.08279v1 Announce Type: cross Abstract: Agent memory systems must discard stored information when their history exceeds a fixed token budget. Existing budget-accuracy frontiers quantify the...
arXiv AI
6d ago
When Stored Evidence Stops Being Usable: Scale-Conditioned Evaluation of Agent Memory
The paper introduces a scale‑conditioned evaluation protocol for agent memory that measures how usable evidence degrades as irrelevant sessions accumulate. It logs memory trajectories and reports four diagnostics—budget‑compliant reliability, tail memory‑call burden, failure‑regime decomposition, and the usable‑scale boundary. Experiments on LongMemEval and LoCoMo show that reliability loss varies across memory interfaces and agents, highlighting that scalable‑memory claims must be conditioned on agent, interface, scale range, and interaction budget.
arXiv AI
Oct 21
ECHO: An Auditable Dual-Path Memory System for Long-Horizon Agents
arXiv:2608.21755v2 Announce Type: replace Abstract: Long-horizon agents require more than retrieval over stored conversations: they must preserve source evidence, revise outdated knowledge, and disti...
arXiv AI
6d ago
CALM: Conflict-Aware Lifecycle Memory Governance for Long-Horizon Language Agents
CALM is a conflict‑aware lifecycle memory governance framework designed for long‑horizon language agents. It transforms heterogeneous records into evidence‑grounded memory cells, updates their validity and lifecycle states using support, conflict, and dependency signals, and admits query‑appropriate evidence through training‑free or learned gating based on temporal applicability. Experiments on LongMemEval‑S and LoCoMo with Qwen and GPT backbones show that CALM outperforms state‑of‑the‑art baselines, improves evidence selection, suppresses erroneous memories, and preserves superseded records for accurate earlier‑state reasoning.
arXiv AI
Sep 11
What Should an Agent Forget? Separating What Is Stored from What Is Used
The paper introduces RD-Forget, a training‑free framework that separates what a persistent language agent stores from what it uses at answer time. It keeps a source archive of all observations while a query‑conditioned memory view filters evidence relevant to the current question, using a frozen language‑model curator to group facts into semantic slots and preserve multi‑hop relations. The approach employs rate‑distortion principles to stay within a memory budget and demonstrates improvements across conversational memory, knowledge updating, fact consolidation, long‑context reasoning, and personalization tasks.
arXiv Computation and Language
Sep 1
Agent Zero Memory: Provenance-Aware Long-Term Memory for LLM Agents
Agent Zero Memory is a provenance‑aware long‑term memory system for large language model agents that distills user interactions into three parallel memory structures: an episodic timeline, an associative entity‑event knowledge graph, and a semantic, citation‑locked hierarchical documentary memory. Retrieval is performed via an intent gate, source router, and concurrent searches across the three systems, producing integrated, cited answers that exclude fabrication and require evidence the reader has opened. The system achieves state‑of‑the‑art performance on LongMemEval (95.60%) and LoCoMo (93.60%) while offering a favorable accuracy‑cost‑latency trade‑off across multiple backbone LLMs.
arXiv AI
Sep 10
When Does Memory Help? A Cost-Aware Evaluation of Long-Term Memory in Tool-Using LLM Agents
The paper introduces MERIT, a benchmark that evaluates the marginal benefit of long‑term memory for tool‑using large language model agents while explicitly accounting for cost. MERIT provides episodic tool‑use tasks across three domains, verifies dependence on earlier‑episode facts, and measures memory operations in tokens and dollars. Experiments on GPT‑4.1‑mini, Claude Haiku 4.5, and Claude Sonnet 5 show that memory can significantly improve task success, but its utility varies widely across models and memory implementations, and full replay is rarely cost‑effective.