An agent that inherits six one-line memories may pull at most one archived source record before acting; a directive written into the store can steer that choice: a pointer to the record, a criterion that identifies it, or both. Across twelve registered studies on one instrument lineage (14,760 attempts) we measured where the request goes under each form. On six direct-provider models a length-matched criterion exceeded a bare id by +35.0 points [+31.2, +38.8] (Study D); the contrast failed its registered superiority rule on a nine-model OpenRouter-served panel (Study E). Appending the id cancelled the criterion on three Claude models (Opus 5: 40/40 to 0/40; Study F-x); six byte-matched edits gave each exact string its own effect (Study G), and a re-run at eighty runs per cell left fifteen of thirty replication contrasts within the margin, fifteen unresolved and none beyond (Study G'). A ratification line (+96.0 points on Opus 5) and a budget of two credits restored the target on all three (Study J); across five criterion strings the suffix's cancellation held for four of the five wordings on Opus 5 and all five wordings on Fable 5.1 (Study H2); in a second store every model followed the criterion (Study H1). Continued into a decision, the criterion moved the choice toward the current record (+100.0 points, Opus 5) and away from it on Fable 5.1 (Study I). A one-character plan pointer's effect (+78.0 points; Study B, after a correction of its first repository report) returned the same verdict under a prospectively registered re-run (+81.7 points; Study B'). All results are descriptive effects of exact edits on fixed panels with registered intervals and no mechanism claim.
LLM agents that cache recovery suggestions from API errors can skip re-derivation in later episodes, spending fewer tokens and fewer model calls on constraints they have already learned. Server-side data drift turns those cached fixes into silent failures, and the usual remedy, re-deriving on every episode, gives the savings back. We introduce invalidation contracts, a protocol layer that attaches version stamps and cacheability hints to every recovery suggestion so the client can evict stale entries without trial and error, and keep the rest. The contract decomposes realized savings into two independent factors: validity, the fraction of cached suggestions that remain correct after a drift event, and compliance, the fraction the planner applies on the first attempt. Validity depends only on the protocol and is vendor-independent. Compliance depends on the planner model: identical wire bytes yield 100% first-try compliance on Claude Haiku 4.5 and 11% or below on Claude Sonnet 5, which exhibits input-schema conservatism, refusing fixes that add fields the original request did not contain. We evaluate across seven models, three serving paths, two domains, and approximately 9,400 episodes. Row-level invalidation raises compliance by 0 to 66.7 percentage points across the seven models, 55.6 to 66.7 on three, and recovers 29-33% of baseline token cost on four of seven models, while table-level invalidation destroys co-located entries and drops post-drift first-try rates to 0% on five of seven. Eviction precision is 1.00 at row granularity on every model under the row-level oracle of Section 4.1. The contract adds 15% to response payload. Version-stamp validity is deterministic by construction and produced identical results across every model and serving path, with zero contract failures in the entire evaluation.
Knowledge graphs have been proposed as a structured alternative to flat retrieval-augmented generation for long-term agent memory, on the assumption that representing conversations as entities and relations improves recall. We evaluate that assumption directly. Our framework extracts each conversational turn into typed nodes and attributed edges, answers questions from a two-hop subgraph, and periodically prunes nodes that score low on a weighted combination of recency, access frequency, degree centrality, and age. On LongMemEval, the graph does not outperform a flat vector baseline at a matched candidate-generation budget of five retrieval roots: token F1 is $0.417$ against $0.468$, and a paired bootstrap over 500 questions gives $Δ= -0.050$ (95\% CI $[-0.085, -0.016]$). The gap is widest on questions that require recalling a specific prior assistant turn, where judged correctness falls from $0.911$ to $0.607$, suggesting that decomposing a turn into entities discards the surface form these questions depend on. The forgetting module is more successful. Applied once to a persistent 27{,}021-node graph, it removes 9.8\% of nodes and 9.5\% of stored bytes; token F1 is unchanged ($+0.001$, 95\% CI $[-0.015, +0.016]$) and judged correctness falls by $1.6$ points, with the 95\% interval bounding any loss at $3.8$ points ($[-0.038, +0.006]$). Because our extractor is a single small model evaluated on one benchmark, these results characterise this extraction-based pipeline rather than graph-structured memory in general. Code: https://github.com/skhanzad/Selective-Amnesia
Reliable handling of unanswerable questions (UAQs) is critical for trustworthy LLM-based agents. Although memory is widely used in agent systems, its role in reliable UAQ handling remains unclear. We present a systematic study of agent memory for UAQ handling under a unified agentic RAG framework, evaluating four representative memory methods across three UAQ-related datasets and two base models. We find that memory can improve UAQ performance in some settings, but such gains are selective rather than universal and remain fragile under dataset shift. Interestingly, cross-model memory reuse is often more feasible than cross-dataset transfer, suggesting that shifts in answerability patterns pose a greater challenge to memory reuse than changes in the base model itself. We further find that UAQ gains are more strongly preserved through decision guidance than through trajectory shaping, and that memory effectiveness depends strongly on representation. In particular, procedural and rule-based memories often provide the most reliable support for UAQ handling, while memory composition is most effective when procedural guidance is combined with complementary behavioral signals. Overall, our findings suggest that reliable UAQ memory depends less on storing larger amounts of experience and more on preserving transferable behavioral guidance.
Provenance links keep the evidence behind an inherited belief reachable; an agent with a verification budget must still choose which links to inspect. We study a consolidated memory that states a decision constraint and whose source record has since been superseded by a record that withdraws it: provenance is immutable, the current record has changed, and the memory is stale. In a controlled six-memory scenario with a budget of two records, sixteen language models rarely re-verified a constraint that read as settled: they inspected its provenance path in about one episode in five and, once the constraint had been superseded, produced stale-consistent decisions in 77.3%, 74.7% and 74.7% of episodes across a primary run, a replication and a held-out domain. Re-assigning one of the same two slots to the critical path removed most of them: +74.0, +72.7 and +61.3 points (positive in every model), +80.7 in a prospectively frozen interleaved replication with a repaired non-critical control, and +62.0 on a panel of 10 models from 9 organisations; a corrected re-run of the held-out scenario gave +73.3. The forced-critical policy uses experimenter knowledge of the critical path: it quantifies how much stale-decision risk the same budget can recover and is not a scheduler. Two further deposited experiments locate the failure and a remedy: in this store the constraint's path is selected in 17.0% of episodes at two slots and 88.7% at four of six (above uniform allocation), and at two slots a one-sentence, target-blind rule (prefer memories that state a limit on a candidate direction) moved the agent's own allocation onto the constraint's path and recovered the oracle contrast on decisions (+89.3 points) where that constraint limits the tempting action, while a content-free freshness cue did not materially redirect allocation and a content-matched control rule changed neither selection nor decisions.
Memory-augmented agents maintain compact user profiles throughout extended conversations, enabling personalized and consistent responses without the need to process the entire dialogue history. The quality of these user profiles relies on the underlying memory management strategy: at each step, the agent must determine what to retain, compress, or discard. However, existing methods typically employ a static, one-size-fits-all strategy established before training. In practice, the optimal memory decision is inherently user-specific and dynamically evolves alongside policy optimization. To address this, we propose \textbf{HiPS} (\textbf{Hi}erarchical \textbf{P}ersonalized \textbf{S}trategy), a framework that decouples memory management into a globally shared foundation and a user-specific adaptive tier. Specifically, HiPS employs \textbf{Universal Strategy} to extract shared principles from cross-persona trajectories, alongside \textbf{Persona Delta Distillation} to generate tailored rules for users whose behaviors diverge from general patterns. \textbf{Cross-Level Rule Flow} dynamically calibrates their boundary by promoting broadly validated personal rules and demoting contradicted global ones. The architecture establishes a co-evolution loop where a mechanism guarantees that all strategy refinements are anchored to task outcomes. Extensive experiments demonstrate consistent improvements over memory-augmented baselines.
Long-term language-model agents accumulate memories across interactions, but their retrievers typically do not accumulate retrieval experience. Semantic retrieval is efficient, but embedding similarity does not always reflect whether a memory contains evidence relevant to the current query. Large language model (LLM) rerankers provide stronger query-conditioned relevance scores, yet stateless reranking repeatedly scores a large candidate pool and discards these scores after each query. We introduce EARM, an experience-amortized reranking framework that treats previously acquired LLM relevance scores as reusable retrieval experience. EARM stores sparse query--memory relevance scores in an online matrix, learns their shared structure through causal matrix completion, and combines a small set of newly observed scores with estimated scores to rerank the remaining candidates. The scoring budget decreases as experience accumulates, changing LLM reranking from a repeated per-query expense into a retrieval capability learned over an agent's lifetime. Experiments on long-term conversational memory show that mixed observed-and-estimated reranking improves answer accuracy over semantic retrieval by up to 6.62% and remains effective when only 17.5% of candidates receive direct LLM relevance scores, thereby substantially reducing the inference overhead of LLM reranking. These results motivate a broader view of agent memory: a long-lived agent should remember not only past content, but also how that content has proved useful for retrieval.
Large language model (LLM) agents can induce skills from completed tasks and reuse them later to grow more capable with experience. In practice, induced skills may transfer unreliably and can even harm the agent that retrieves them. When agent-induced skills transfer reliably across tasks remains an open question. We conduct a comprehensive and controlled study of how the way skills are induced shapes their transfer across tasks. Specifically, we compare task-level with subtask-level skill induction and text with code skill formats, the two axes along which existing methods differ. Task-level skills mostly reduce the agent's performance below its no-memory baseline while subtask-level skills raise it above on average, and text skills transfer better than code skills. To further understand our findings, we examine two complementary properties of the induced skills: specificity, which measures how closely a skill matches real tasks, and abstractness, which measures how evenly its relevance spreads across tasks. Neither property alone predicts task success, but their combined effect does, which we propose as a skill utility score. The score correlates consistently with task success when skills are transferred, and subtask-level and text skills score higher. Computing skill utility only needs the skills and task descriptions but not any task execution, so our score serves as a practical diagnostic of a skill memory before any new task runs.
Memory is a key capability of LLM agents. Persistent memory extends this across sessions---enabling recall, revision, and personalization. Yet its multi-stage pipeline (ingestion, retrieval, filtering, generation) makes failures difficult to localize: end-to-end evaluation reveals that an error occurred, but not which stage caused it. Existing evaluations often report aggregate performance without paired statistical comparisons, slice-level non-regression checks, or stage-level diagnostic traces. We propose D$^2$ACCI (Diagnostic-Driven Artifact-based Closed-loop Controlled Iteration), a dual-loop protocol whose outer diagnostic gate promotes, feature-flags, or rejects memory interventions based on paired evidence, protected-slice monitoring, and trace-level localizability. We further introduce DCR, a graded observability metric that measures whether failures remain localizable, and D$^2$ACCI-Eval, a reusable artifact for gate replay. We instantiate the protocol in MemStack and evaluate on three public benchmarks, achieving 93.59% on LoCoMo, 90.93% on LongMemEval, and 57.20% on PersonaMem-V2. Five paired ablations show that supplement extraction, session-memory retrieval, and Forget Guard yield statistically significant gains (+1.9 to +3.7pp, all p $\le$ .003). In contrast, BM25/RRF is retained as a monitored feature flag---a distinction invisible to aggregate-only evaluation. A diagnostic audit shows enriched traces substantially improve root-cause agreement over result-only relabeling. Diagnostic artifacts reach 98--100% DCR@3 versus 0% for results-only logs. These results establish that robust memory-system iteration demands traceable, statistically grounded, and regression-aware evidence---exactly the gap D$^2$ACCI fills.
Lauri Lovén, Jaakko Sauvola, Jukka Riekki +1cs.DC cs.AI cs.MA
Autonomous agents share a transport and can call each other's tools, but they cannot share what they know: no protocol lets two agents' memories reconcile a fact phrased two ways, link related facts held apart, or reconcile contradictory knowledge without silently discarding either claim. We present MELD, a self-managing coherence mechanism for a federation of agent memories whose run-time model is the knowledge graph itself. Each brain admits every incoming claim through a five-outcome procedure (insert, merge, relate, conflict, or reject), decided from three signals (scoped claim-key identity, embedding similarity, and a natural-language-inference verdict) under context and freshness gates, and acting through exactly one auditable, authenticated Patch, the only object that mutates state. A binding onto standard publish/subscribe transport with a per-claim status CRDT keeps sovereign brains coherent in claim status without a coordinator: self-healing after partitions and under lossy routing, and self-protecting against silent rewrite by a peer, under a benign-fault model. MELD does not adjudicate truth; a detected contradiction is preserved for later adjudication, never silently resolved. On HotpotQA distractor, distributed merge is recall-non-inferior to a centralized store under a pre-specified equivalence test and recall-superior to naive union at about 11% less live storage; the merge classifier separates at AUC 0.968 with a 0.013 false-merge rate on adjudicated candidate pairs; the status CRDT reconverges in 30/30 real partition-heal trials where last-writer-wins manages 11/30; and semantic routing delivers about 3x fewer messages at matched recall. We evaluate on a real computing continuum spanning an operator-grade 5G edge, national HPC, and a local tier, with empirically calibrated thresholds.
Most agent-memory benchmarks test post-hoc recall, whereas MemoryArena evaluates whether memory supports interdependent, multi-session task completion. We compare MemoryLake, a structured multi-track memory backend, with Mem0, text-embedding-3-small vector RAG, and a long-context control across all five MemoryArena domains. The systems share the same agent framework, requested gpt-5-mini model alias, task samples, and scoring code; the memory integration is the intentionally changed component. Because each backend bundles write, retrieval, consolidation, budgeting, and prompt-assembly choices, the study is a matched system-level comparison, not a representation-only ablation or a cost-matched experiment. On the shared evaluation sets, MemoryLake has the highest observed success rate (SR) in mathematics (9/40), physics (12/20), and progressive retrieval (4/20). Every system has zero SR in travel planning, and web shopping yields a single bundle-level success (long context, 1/150); MemoryLake ranks third on both the travel soft process score and shopping step match. Following MemoryArena's suite-level convention, a post-hoc equal-weight average over the five SRs is 20.5% for MemoryLake versus 13.6% for the best comparator. These are point estimates: sample sizes are modest, confidence intervals overlap, and we do not report paired significance tests. A separate MemoryLake-only run over all 221 progressive queries yields a failure-counted SR of 26.7% (59/221) and is not a baseline comparison. The results support a workload-dependent view of memory backends and an observed lead among the four evaluated systems on the shared sets; they do not establish benchmark-wide state of the art or a causal advantage of representation structure.
Agent-memory systems increasingly buy retrieval quality with structure, transforming raw conversation histories into summaries, embeddings, trees, or knowledge graphs before any question is asked. We ask how much of that benefit comes from the structure itself, rather than from competent retrieval over the raw history. We present ReFind, an agent-controlled search interface that builds no semantic structure at all: it leaves the conversation archive unmodified, indexes it lexically at turn granularity, and combines a generic iterative keyword-search loop with four chat-native controls grounded in empirical refinding work: session-aware rank fusion, local context expansion, temporal narrowing, and skipping already-inspected sessions. A separate reasoning stage answers from the collected evidence. Across a broad suite of conversational-memory tasks (single- and multi-hop QA, event ordering, and fact consolidation), roughly 2,800 questions on precise-retrieval and fact-tracking capabilities evaluated under the incremental multi-turn setting of MemoryAgentBench, ReFind attains the highest mean accuracy (58.2) of any system compared, above the strongest graph- and tree-based memory systems (HippoRAG 2, 53.2), all under a GPT-4o-mini backbone matched to every reused baseline. Controlled comparisons to single-shot BM25, a matched generic-agentic BM25 control, component removals, and agentic dense/hybrid variants separately support the roles of agent control, chat-native controls, and lexical retrieval. On LongMemEval-S/M, the same interface reaches 93.2 +/- 3.3 and 89.3 +/- 6.0 with GPT-5-mini. The results indicate that for precise, evidence-grounded questions over chat archives, much of the benefit credited to elaborate memory structures is recoverable by giving an agent controllable search over the unmodified record, with no LLM-based index construction at all.
Despite the wide deployment of memory in large-model agents, there is no unified formal account of what a memory is or when it is optimal. This paper takes a first step toward this account. The central idea is that memory is a basis, knowledge is its span, and answerability is a coverage problem: an agent stores events extracted from a material; a generation operator turns any event set into the knowledge it entails; and a query is answerable exactly when some single item in the span covers it. The optimal memory is then the capacity-constrained maximizer of expected coverage, and its value traces a utility--capacity frontier, the common yardstick on which memory systems can be compared. Next, we consider noise in the memory and discuss coverage versus precision under it: a memory may store false claims, so the write policy must infer the truth of what it stores. Drawing an analogy with biological memory, which is formed continuously through ongoing experience, we formalize the continual agent-memory problem in a sequential MDP that covers multiple levels, where memory is the state, writing is the action, and the utility settled at query time is the delayed reward that drives learning. To make the framework concrete, we instantiate it on Homer's \emph{Odyssey}, turning the frontier, the compression zone, and the divergence of coverage from precision into concrete numbers. Finally, we position existing systems within the framework, making ``how good is a memory'' measurable and recasting the open problems of constructing and learning agent memory as concrete research questions.
Long-horizon agents accumulate trajectories spanning hundreds of interleaved reasoning, action, and observation steps, where answering a query may depend on evidence buried far back in the history. External memory stores such trajectories as structured representations, yet each structure provides a distinct and incomplete view. Existing multi-memory systems either read a fixed set of structures for every query, inflating context and introducing noise, or route each query to a single structure, preventing the composition of complementary evidence. A controlled analysis on AMA-Bench shows that the optimal memory configuration is typically neither a single structure nor the full union, but a tailored composition of multiple structural memories that varies with query and task demands. Motivated by these findings, we formulate structure-level dynamic selection: selecting and fusing a query-adaptive subset from a library of specialized memory structures. We propose MESA (a Multi-structure Evidence Selection framework for long-horizon Agent), which builds five complementary structure views of each trajectory and learns from end-to-end answer-level feedback to select and fuse a query-specific subset for a frozen answer model. To learn under this weak supervision, MESA employs harness optimization with prior-guided search and UCB-guided scheduling to balance exploration and exploitation. On AMA-Bench, MESA outperforms the strongest baseline by 8.5% while using 41% fewer evidence tokens than the all-structure alternative.
We present SuperLocalMemory 4.0, a governed, local-first memory operating system for AI agents, unifying multi-channel retrieval under reciprocal-rank fusion, bi-temporal recall, multi-scope isolation, role-based access, verified erasure, and a hash-chained audit trail. A reliability spine governs the primary write path: generation-fenced admission, verifiable memory transactions with per-projection apply, verify, compensate and erase owners, and hash-checkable completion manifests. Eleven fault-injection scenarios, each repeated 200 times, upheld 2,199 of 2,200 scoped component properties. This version leads with a negative result. Ten mechanisms here were implemented, reachable on a live call path, and ineffective at their final connection. Implemented, reachable and effective are three different questions, and the third requires an oracle independent of the mechanism under test. We contribute two mechanical invariants that supply one: a prior-distance assertion over Bayesian learners, and a join-liveness assertion over schema-guarded paths that reports where a guard's missing data resides. A three-arm ablation varying only the recall session-identifier namespace moves no posterior with the defect present and every instantiated arm with it absent, while a negative control that writes every ticket but supplies no engagement settles nothing. We withdraw the previous version's governed write-envelope overhead figure: the two paths it differenced are not comparable. Timing the envelope in place gives an 11.0 ms governed write of which the envelope is 70.6 percent, but the generation fence costs 1.9 microseconds and the obligation ledger 42 microseconds. The cost is durability, not governance.
Agent memory systems compress what they store, and compression is built to drop qualifiers, so a claim's epistemic standing tends not to survive being written to memory. We ask what governs whether it does. Matched notes carry the identical claim and identical stance and differ only in where that stance sits; one model compresses both under the same budget among the same filler notes, and a blind reader that never sees the condition scores the result. Across 60 claims in seven registers, writing the stance as a labelled field rather than a bracketed aside raises retention by about 15 points on two models (37 claims to 2 on one, 30 to 8 on the other; permutation p=0.00005), and a pre-registered replication on Haiku, its prediction and decision rule committed before the run, gives +15.6 points, 38 claims to 1. Ablating the format on both models gives the same net effect from different parts: labels help on both (+9.7 and +12.8) and length helps on neither, but wording the stance as a full sentence is the largest component on one model (+12.5) and worth nothing on the other (+0.6). Either model alone would have licensed a confident and different mechanism, so we claim only the intersection: make the stance explicit, not merely longer, and expect the best way of being explicit to depend on the model. A deterministic readout with no model reproduces the two-cell direction and five of seven ablation contrasts, but not length or labels, which we therefore do not claim on one instrument. Fifty hand labels (kappa=0.75) agree on direction; we print their seven disagreements in full. We also report nine withdrawn claims, three of them former title claims of this paper.
Computer-use agents pay full frontier inference to re-derive routines their user has already performed, because an agent's memory today records what the user said, not what the user did. We compile passively captured screen activity into agent memory with a deterministic, zero-model pipeline: it segments a local capture stream into typed activity frames, bounded episodes carrying application, site, timing, input volume, and evidence pointers back to the raw rows, with no model in the loop, so the output is byte-identical, cacheable, and mechanically auditable. On one professional's single-user corpus of 128,756 frames over 51 active days, the compiler reduces a day of raw capture to a prompt-ready context block 86x smaller in 68 ms, and an agent reading that block answers questions about the day at 98.4% accuracy (Wilson 95% CI 91.7-99.7%) against an independent oracle, versus 66-80% for an LLM summary of the same capture, a mid-tier model reading the block matching a frontier one. The same compiler doubles as a demand-side cost instrument. Read off passive, pre-delegation human activity rather than agent rollouts, it supplies two parameters that agent-cost models assume but, to our knowledge, have not measured: the Routine Overhead Ratio R and the routine recurrence h. We report first values of R, a modeled upper bound, at 60-343x, and a delegable recurrence of 9.0% in-sample and 7.7% out-of-sample, for a realistic all-fleet token ceiling near 8%; a compiled routine replays deterministically with the model out of the loop, demonstrated live at zero model tokens on a guard-matched hit. Schema, compiler, and evaluation harness are open.
GUI agents must remember both useful experience from earlier tasks and unfinished progress in the current interaction. Latent memory offers a compact solution by compressing multimodal trajectories into a few continuous tokens. Existing methods, however, usually map each trajectory to one fixed memory block and train it mainly through next-action supervision. This creates three practical problems: important details may be lost during compression, the same memory block must serve different decision stages, and irrelevant retrieved trajectories may still mislead the agent. We introduce FocusMem, which separates these responsibilities within a compact latent-memory interface. A role-aware content basis encourages episodic memory to retain reusable experience and working memory to retain task progress. A state-conditioned readout generates a decision-specific view of the same stored evidence, while a lightweight trust gate can suppress memory blocks that appear irrelevant to the current step. All components are trained while the GUI policy remains frozen. Across five GUI-agent benchmarks, FocusMem consistently outperforms a fully matched action-only fixed-memory baseline and prior latent memory adaptations. Further analysis shows that semantic and functional supervision preserve complementary information, state-conditioned readout is more robust as surrounding trajectory context grows, and the trust gate reduces the harm caused by injected irrelevant episodic evidence. These results show that effective latent memory depends not only on compressing past interaction, but also on what is retained, what is exposed, and what is allowed.
Persistent memory helps long-term agents retain knowledge, yet a single update error can repeatedly distort future retrieval and reasoning. Most existing systems reduce memory updating to a binary Write/Hold decision, which cannot distinguish whether new information should be added, ignored, used to revise an outdated belief, rejected as unreliable, or deferred for verification. These choices may share the same binary label while producing fundamentally different memory states. We introduce TARL, a memory state update framework that maps each statement to one of five executable actions. TARL identifies the affected memory, resolves its temporal scope, compares source reliability, and updates accepted, pending, and rejected ledgers. It is further trained by comparing the memory states produced by alternative update operations, encouraging the model to select the operation that leads to the correct result. We also introduce TARL-Mem, a benchmark with fine-grained action labels and next-state targets. Across in-domain, cross-source, temporal, counterfactual, and sequential evaluations, TARL improves action prediction and state recovery, reduces memory pollution, preserves conflicting evidence, and limits cumulative corruption. The complete model implementation is provided in the supplementary material.
Yi Yang, Zhennan Chen, Yihong Zhuang +5cs.LG cs.CL
Learning-based memory systems for self-evolving LLM agents face two tightly coupled challenges. First, trajectory-indexed utilities grow with the interaction history, thereby dispersing limited feedback over an ever-expanding state space. Second, because trajectory-level rewards are jointly assigned to co-retrieved memories, irrelevant experiences may receive misleading utility updates and consequently enter the memory-reward trap. To address these challenges, we introduce Reduced-Order Memory Reinforcement Learning (RoMeRL), which represents the growing trajectory-indexed utility space using a fixed-dimensional per-task memory state factorized by outcome polarity and memory dynamics. RoMeRL incorporates new experiences through a fixed set of semantic coordinates whose contents are updated or replaced over time, thereby concentrating feedback over a bounded utility support. Theoretically, we show that this reduced-order parameterization increases the average feedback received by each utility coordinate and characterize the steady-state occupancy of erroneous coordinates under a generic coordinate-transition model. Empirically, across ALFWorld and LifelongAgentBench, RoMeRL improves task performance, reduces the Cold-Q ratio by 80.0%, increases feedback density by approximately 6.0 times, reduces the maintained memory size by 84.4%, and cuts LLM calls by 21.1%. These results show that reduced-order utility states support efficient self-evolving agent memory while limiting persistent reward contamination. Code is available at: https://github.com/YOUNG-fnxm/RoMeRL
Persistent memory lets long-running large language model agents reuse information across sessions and tasks. Yet errors in writable memory can persist and corrupt future behavior. Existing systems improve storage and retrieval, but they do not provide a transaction boundary for reliable updates and recovery. We therefore propose MemTxn, a governance layer outside the answer model. MemTxn verifies whether an update is supported by its source. It also selects the visible version when facts conflict and restores the application-visible state after a fault. The system uses Ordered PatchTest to validate writes, a Temporal Resolver to select versions, and a durable snapshot journal to recover state. On an item-disjoint audit, MemTxn accepts all 60 supported originals and rejects all 179 hard negatives. Under persistent multi-key faults on LongMemEval-S and LoCoMo states, it restores the complete declared active map without knowing the actual physical write set. On MemoryAgentBench FactConsolidation, MemTxn achieves the highest average F1 across all twelve answer-model configurations. It outperforms Dense by 17.06--24.07 points in five representative settings.
LLM agents increasingly rely on long-term memory to support multi-session interaction and personalization. However, existing agent memory systems are designed around forward-only evolution, continuously accumulating, consolidating, and overwriting knowledge, with no principled mechanism to inspect, version, or revert prior states. This makes agents brittle under corrections, concept drift, and memory corruption, particularly after they have already been exposed to subsequent information. We present ChronoMem, a semantic version-control layer for agentic memory integrated into the production-ready, open-source Agent Development Kit by Google. ChronoMem commits whole-memory snapshots at each memory write, maintains structured version histories, and supports natural-language rollback requests by mapping undo intents to concrete historical versions through hybrid lexical and semantic retrieval, rank fusion, and reranking. We further introduce a post-exposure evaluation protocol that tests whether an agent can behave counterfactually after rollback by answering queries and summarizing history as if future updates had never occurred. On long-horizon conversational benchmarks augmented with evolving memory states and rollback tasks, ChronoMem substantially improves rollback-consistent question answering and history summarization relative to prompt-only and retrieval-only baselines, while achieving strong performance in semantic version selection. To our knowledge, ChronoMem is the first open-source system and benchmark for systematic semantic global memory rollback in LLM agents.
Deployed LLM agents increasingly keep their long-term memory as a filesystem: a directory tree of markdown files that the agent itself reads, writes, and reorganizes through generic file tools. Yet research has largely passed over this medium: prior systems design bespoke memory representations and study retrieval over them, leaving the default's two working assumptions untested: that an agent can keep a growing store organized as memories accumulate, conflict, and go stale, and that this organization pays. We present the first systematic exploration of filesystem-based memory for LLM agents. We formalize the setting as three roles around one memory filesystem: a management agent integrates and organizes incoming content, a search agent answers queries with cited sources, and an execution agent supplies task trajectories that are distilled into skills, unifying declarative memory and skills in a single store. Across long-conversation benchmarks and embodied tasks, we vary memory shape (agent-organized hierarchy, verbatim dump, chunk retrieval), stream scale, tool harness (sandboxed shell, memory-tool-style functions, varied search tooling), and the strengths of the management and search agents, tracking answer quality, cost, and store health as memory grows. What organization reliably buys is search economy: organized stores roughly halve retrieval cost where material is large. Today's agents, however, fall short of the default's promise: in our growth study, organization erodes for all but the strongest management agent, and no agent we measure converts organization itself into better answers. And the model is not the only lever over a store's shape: changing the tool set alone reshapes the store as strongly as swapping the model. The study turns the filesystem default from an assumption into a design space for agent memory.
Long-term memory systems store what a user says in an external store and retrieve it when a related query arrives. This interface rests on an assumption so natural that it is rarely stated: a memory that is needed will resemble the query that needs it. World knowledge breaks the assumption. A tree-nut allergy should change the answer to a macaron request through their almond-flour ingredient, yet the two texts share no cue a retriever can see. We call this failure mode the implicit-association blind spot and introduce InMind, a 125-task, expert-verified benchmark spanning ten life domains, with 113 tasks grounded in citable public sources. Its paired controls separate three explanations that existing evaluations conflate: the fact was never stored, the model lacks the bridging knowledge, or the fact was stored and never surfaced. The verdict is clean. With the decisive memory placed in context, the backbone answers 84.0 percent of indirect queries; when the same memory must be retrieved, six vector, graph, and agentic memory systems reach at most 14.4 percent, even though they recall the same facts on demand at up to 100 percent. An embedding with eight times the dimensionality raises answer-blind target recall for every system yet leaves the gap essentially intact. A minimal diagnostic probe that keeps memory visible before the query arrives recovers most of the gap, locating the failure in the query-conditioned interface itself and pointing to routing, deciding which facts must stay visible, as the open problem InMind is built to score.
LLM agents increasingly coordinate through persistent shared memory: one agent's write becomes another agent's premise, and eventually a tool call with real side effects. Current agent memory systems treat every accepted write as immediately actionable truth, so a polluted tool result, a stale update, or a teammate's half-finished note can silently drive an irreversible action. We argue that a memory write is not a belief commit. We present MemTX, a transactional belief-commit protocol. Each record carries evidence, permissions, provenance, and validity. Writes are staged inside snapshot-isolated transactions and admitted by a validate-and-commit pipeline, irreversible tool calls are gated on in-flight belief state, and retracting a belief triggers typed cascading repair of its derived records and tool side effects. Two invariants, action-safety gating and cascade-repair completeness, are machine-checked by property-based testing and bounded exhaustive enumeration of 5.5 million protocol states, with zero violations. Across five backbones from three model families, MemTX leads all eight baselines with paired-McNemar significance on four backbones and statistically ties the best baseline on the fifth and strongest, while remaining the only method with zero downstream harm on every backbone. Backbone capability does not substitute for commit discipline.
Benchmarks for LLM-agent memory typically generate conversations first and extract answer keys afterwards -- with documented label-error and contamination problems -- and they overwhelmingly measure short interaction histories. We invert the pipeline: a seeded life-script sampler emits facts with validity intervals, volatility classes, and source channels before any text exists; an LLM renderer writes chat and email from per-event fact manifests; a fidelity verifier confirms every planted fact; and questions are instantiated mechanically from the script, so gold answers are script-valid by construction and separately validated for answerability. The synthetic, fictionalized corpus (~380 questions, 15 types) embeds features absent from the benchmarks we survey: per-fact validity intervals, sent/received trust distinctions, injection probes in a benign harness, and as-of-date question sets. Benchmarking five memory architectures against a no-memory control (fixed answerer, versioned LLM judge, three replicates, two horizons), we find backend rankings invert with history length: the budgeted curated-map memory that leads at three weeks loses recall of evicted content by nine weeks (96% to 72%) while a provenance-typed graph rises to 90%; the inversion is positive for all six users under complete cross-family re-judging (exact p=0.031). A full-rendered-history baseline ties or exceeds the best memory system at the short horizon but shows no judge-independent advantage at nine weeks, at about twice the read cost. Write-stage quality strongly correlates with downstream quality (weakly-written facts fail 24% vs 2%), and injection resistance tracked whether provenance boundaries survive representation. A layered architecture performs best among the memory systems in both regimes (96.8% short-horizon) and is released as Veracium, an open-source library, with the corpus generator and harness.
Production AI agents' failures are less often due to an inability to reason well and more often because they cannot manage what is in their reasoning context: conversation histories, large prompts, large tool definitions, and ballooning tool outputs. Agents drown in their own accumulating history while paying a token cost that grows every turn, producing missing recalls within and across conversations. The incumbent response treats this as a storage-and-retrieval problem. We argue that framing is too narrow. Actively managing what an agent holds in mind is a lifecycle, not merely a store: it spans deciding what to remember, extracting and structuring it, choosing the right store per data type, consolidating and forgetting while preserving provenance, deciding what is relevant now, anticipating what is needed next, and compacting context to a budget without losing what matters. In serious production this operates not over a single user but across an organizational scope hierarchy. We name this discipline Agentic Context Management (ACM) and decompose it into five primitives: architecting, ingesting, scoping, anticipating, and compacting & consolidation. We then make the economic case: naive context accumulation grows token cost quadratically in conversation length, crude summarization buys linear cost at the price of an accuracy cliff, and only validated compaction achieves linear cost with preserved fidelity. We describe a reference implementation, Maximem Synap, that realizes the five primitives as a multi-tenant service and reports 92% on LongMemEval and 93.2% on LoCoMo under the configuration detailed in Section 6. We close with dimensions existing benchmarks do not yet capture, latency, token efficiency, and context-rot resistance, and the frontier of decision-level and organization-level context the category points toward.
While memory systems are essential for agent architectures, pervasive architectural fragmentation restricts systematic research. Existing implementations typically couple different stages of the memory lifecycle, entangle evaluation logic with specific datasets, and provide limited support for the management of heterogeneous memory types. We introduce MemTools, an interoperability research framework that decouples memory system components from their underlying deployment environments. MemTools standardizes the memory lifecycle through declarative data contracts, enabling the interchangeable assembly of components across different systems. It orthogonally separates benchmark datasets from execution protocols to facilitate controlled assessments. Furthermore, MemTools provides a unified computational interface for coordinating symbolic, neural, and multimodal memory representations within a shared runtime. Empirical evaluations on cross-system component integration, evaluation protocol reconfiguration, and heterogeneous memory coordination demonstrate that MemTools enables systematic isolation and analysis of memory design variables. These findings suggest that MemTools provides a practical and extensible infrastructure for advancing principled research on agent memory.
Effective memory is crucial for LLM agents, yet constructing it effectively remains challenging. A memory-construction policy decides what information to extract, store, update, compress, or discard as interactions accumulate. Heuristic memory methods rely on subjective, task-specific rules, which can misalign with downstream objectives and limit cross-task adaptability. RL-based methods, by contrast, learn from task feedback but mainly use outcome- or module-level rewards. These coarse signals indicate task success but cannot identify which intermediate memory contents support the final answer, creating a fine-grained credit-assignment bottleneck. However, constructing such process feedback is prohibitively difficult because intermediate memory decisions lack unique ground-truth targets, while the appropriate credit varies with the agent's uncertain reasoning trajectory and therefore cannot be specified in advance. We propose AttriMem, an attribution-guided process-feedback framework for learning memory-construction policies with RL. AttriMem augments the global outcome reward with local rewards derived from token-level contributions to the final answer. Experiments on long-horizon dialogue question answering show that AttriMem outperforms retrieval-based, heuristic, and RL-based baselines, generalizes across benchmarks and answer models, stabilizes RL optimization.
Agent-memory workloads mix direct factual lookup, relation-chain and current-state reasoning, and broad synthesis over long histories. We describe Supra Cognitive Modes (SCM), an architecture that maps explicit or automatically selected per-query modes to retrieval and synthesis payloads over one shared ingest substrate. A frozen semantic classifier and runtime gates dispatch queries among fused lexical and dense lookup, graph or iterative multi-hop handling, and stratified long-form synthesis. The substrate combines multi-granularity embeddings, extracted triples, fact-version metadata, and optional asynchronous enrichments. We characterize the deployed configuration on three benchmarks: Long-term Conversational Memory (LoCoMo; n = 1,986), MemoryAgentBench (MAB; n = 3,671), and LongMemEval (n = 500). The reference run records 84.87% on LoCoMo factoid categories and 68.61% on adversarial abstention, 61.49% on MAB across two repetitions, and 86.00% on LongMemEval. A repository-backed reproduction produces similar aggregate scores and supports task- and mode-conditioned failure analysis. Raw baseline outputs, aligned end-to-end timing for LoCoMo and LongMemEval, and complete token ledgers are unavailable; stored rows also omit some final runtime decisions. The results characterize one implemented routed configuration and its diagnostic failure patterns, while source inspection verifies the per-query control interface and shared-substrate design. Causal routing effects, efficiency gains, and statistical significance remain outside the available evidence.