As large language models (LLMs) become increasingly capable, the long-term value of AI systems depends not only on solving individual requests, but also on transforming experience and accumulated knowledge into durable, reusable competence. We introduce SimSkill, a self-evolving agent built around the Simulation of Urban MObility (SUMO) traffic simulator. SimSkill identifies capability gaps, generates and solves environment-grounded tasks, verifies solutions through an action--critic loop, and consolidates experience into episodic, procedural, and semantic memory without updating the backbone model. Through autonomous exploration, it builds a reusable library spanning the traffic-simulation workflow. We evaluate SimSkill on two held-out benchmarks with three backbone LLMs and independent artifact-based verification. SimSkill improves verified completion by up to 25 percentage points, while ablations show complementary contributions from procedural and semantic memory. Its benefits remain backbone- and budget-dependent: memory does not improve every model or uniformly reduce inference cost. More broadly, SimSkill illustrates a design paradigm in which natural language preserves and composes computational capabilities, while executable tools and code provide precise and reproducible execution. All code and experimental data are publicly available at https://github.com/qiliuchn/SimSkill-V1.
Multi-agent LLM systems commonly use an orchestrator to decompose a task for a team of workers and then improve through textual reflection. Despite strong empirical results, these systems lack a unified account of coordination, memory improvement, and the role of external verification. We model orchestrator-worker interaction as a bilevel coordination game: under bounded coupling, the workers' local-update game is an approximate potential game whose equilibrium slack is controlled by decomposition quality. We then analyse reflection as stochastic movement over semantic memory states. For free-form reflection, we derive a finite-time upper bound, prove worst-case tightness, and give a positive lower bound under a falsifiable persistent-harm condition. We further prove an information-theoretic impossibility result: no gate that observes only the generated transcript can improve uniformly over text-indistinguishable environments, whereas an environment-grounded gate can. Motivated by this separation, we introduce Stochastic Reflective Memory Ascent (SRMA), which accepts a candidate memory only after a grounded evaluation risk strictly decreases. Under calibration and non-degenerate corrective mass, SRMA converges exactly, geometrically or polynomially; matching constructions show that both rate regimes are order-tight. We also provide confidence gating for stochastic evaluation and re-anchoring guarantees for piecewise-stationary environments. Experiments instantiate these objects with environment-grounded metrics and test the predicted coordination and drift laws. On 500 SWE-bench instances, the complete Kimi-based system resolves 72.2% versus a 70.8% public mini-SWE-agent reference. Code: https://github.com/YihangChen9/Bilevel-Coordinated-Reflection
Data-centric agents repeatedly perform a discovery step before planning or execution: identifying the data objects relevant to a task. Yet successful discovery outcomes are typically discarded rather than reused. We introduce persistent discovery context, a lightweight memory layer that stores prior intent-to-object mappings and reuses them to augment future retrieval. Across three structured data environments, persistent discovery context consistently improves retrieval quality over metadata-only search, remains effective with automatically generated memories, and exposes a reproducible interference failure mode. In lexically sparse domains, memory-only retrieval can even outperform metadata-based retrieval. These findings suggest that discovery outcomes constitute a useful form of reusable context for data-centric agents.
Yijun Chen, Yaqi Zheng, Yanya Li +11cs.CL cs.AI cs.LG cs.MM
Multimodal memory offers a scalable interface for long-video question answering, but existing methods often retrieve captions, frames, transcripts, summaries, or graph facts as isolated fragments. Although searchable, such fragments are not generation-ready: language models must reconstruct cross-modal and temporal alignments at inference time, when context is limited and attribution is difficult. We propose EM^2Mem, an event-centric multimodal memory framework that binds heterogeneous evidence to event anchors during memory construction. Each event-indexed memory cell aligns multimodal records, temporal context, graph-linked relations, semantic facts, and provenance, enabling compact evidence readout over grounded multimodal events rather than modality-specific fragments. Across three long-video QA benchmarks, EM^2Mem improves average accuracy over the strongest memory baseline by 2.0, 2.4, and 3.7 points, improves strict event-level Top-5 evidence recall by 7.0 points, and reduces per-query latency by 4.67 times and total inference tokens by 63.66% (The code will be integrated into https://github.com/zjunlp/LightMem).
Agent systems are commonly described by the model and harness that currently produce their behavior. That boundary is useful for one execution but underspecifies a long-lived agent that may change models, orchestration harnesses, interaction sessions, and host servers while retaining one identity, memory, and executable code lineage. We present a runtime-independent architecture for persistent agents. A continuity-bearing substrate $P_t=(I_t,M_t,B_t)$ contains an architectural identity representation, private durable memory, and a versioned software body. A replaceable deployment binding comprises an execution substrate $E_t=(R_t,H_t,D_t)$, which supplies a reasoner, harness, and host, and a set of interaction surfaces $S_t$, such as chat, API, or user interface bindings. A deployed execution is $A_t=P_t\triangleright(E_t,S_t)$; changing either replaceable layer is migration, not agent creation, when an authorized protocol preserves attributable lineage and transfers continuation authority within a governed deployment boundary. We define six continuity invariants and a quiesce--checkpoint--validate--bind--rehydrate--resume protocol. Enoch realizes the design as a reusable body plus private installed identity, memory, workflow state, and continuation authority, with infrastructure dependencies behind versioned provider contracts. A clean-room run of the frozen public commit passes 833 core tests and 92 provider and library tests executed separately from the core suite; deployments have exercised reasoner-version, interaction-surface, and host-machine substitutions while retaining continuity-bearing state. This evidence supports mechanical substitutability and authorized system continuity, not behavioral invariance or exhaustive pairwise evaluation. The downstream measurement question is whether an authorized continuation still recalls, composes, and enacts its identity.
Arnol Manuel Fokam, Fasseu Sieyondji Akpevwoghene, Edem Fiifi Dawsoncs.LG
The linear recurrent neural network (LRNN) is a simple model for studying how much memory a network builds up as it trains. For uncorrelated inputs, earlier work found that training itself settles the network between keeping the past and reacting only to the present. Real sequences are correlated, and we solve the learning dynamics exactly for correlated inputs. In the solution, keeping the past carries a cost. The whole effect of correlation lands on that cost. This cost reduces to the earlier one when inputs are uncorrelated and grows once they are positively correlated. Three findings follow. (1) Correlation reshapes the course of learning, not only its end. Memory builds, overshoots, and is partly removed, and the settled network keeps less of the past. (2) Memory switches off at a threshold set by one number, how much each input resembles the one just before it. Neither sequence length nor longer-range correlation moves this threshold. Memory is worth keeping only when the task needs the previous input more than the current input already supplies it through correlation with the past. (3) The best network changes too. Zero error demands a feedthrough, a path that passes the current input straight to the network's output and remembers nothing, and training builds it unprompted when given one spare hidden dimension. Our work turns one property of the input into a prediction of whether a network learns memory and explains why correlated data turns recurrent networks into change detectors.
Lars Osterberg, Maggie Wang, Mac Schwagercs.RO cs.CV
While Vision-Language-Action (VLA) models have leveraged internet-scale pretraining and task-focused finetuning to achieve strong performance on long-horizon tasks, they often struggle with non-Markovian tasks that require memory. Existing approaches to memory typically involve additional Vision-Language-Models (VLMs) for long-term memory management, introducing a memory bottleneck and a fractured training pipeline. Conditioning on multiple historical frames can provide the VLA with access to more descriptive features of past scenes, but can degrade performance if frames are chosen at arbitrary, fixed intervals. To address these limitations, we present UniMem, a framework that unifies high-level, multimodal memory and low-level control under one backbone. UniMem employs an event classifier for memory updates, a keyframe encoder for dense spatial memory, and a keyframe caching technique to minimize overhead during policy rollouts. We evaluate UniMem across five simulation and four hardware tasks targeting sequential and spatial memory, demonstrating that our unified, single-model system outperforms fixed-interval image sampling baselines (93.4% vs. 68.2%) in simulation and hierarchical baselines (80.0% vs. 43.5%) in hardware, while offering faster inference and a simple training pipeline for easy adoption. Project website: https://losterberg3.github.io/unimem-vla/
Accurate and responsible medical question answering (QA) is important in healthcare, where complex cases require factual knowledge and nuanced reasoning. Existing medical QA systems, typically based on single-agent architectures and static retrieval, often lack adaptability, persistent memory, and structured decision-making. This work introduces an adaptive memory and reflection (AMR) agentic system, a multi-agent framework in which specialized agents use dedicated memory and reflection-based feedback to retrieve relevant prior cases and improve subsequent reasoning. Complexity assessment routes questions through solo, collaborative, or escalated workflows, while consensus and ethical overseer modules support reasoning consolidation and output review. Evaluation on MedQA and MedMCQA demonstrates strong performance compared with several baselines. Ablation studies show that combining agent-specific memory, reflection, and external retrieval yields the strongest performance. These findings highlight the potential of structured memory and feedback for developing more trustworthy medical agents. The source code is publicly available at https://github.com/mm-air/AMR-Agent.
Proactive interference (PI) is a documented failure mode in large language models in which retrieval of a repeatedly overwritten value degrades as prior overwrites accumulate, mirroring a classical phenomenon in human working memory. Post-training quantization (PTQ) is now the default deployment path for open-weight models, yet its effect on this failure mode has not been tested. We evaluate three precision levels (FP16, INT8, INT4/NF4, via bitsandbytes) across three architecturally distinct instruction-tuned models (Qwen2.5-7B-Instruct, Mistral-7B-Instruct-v0.3, Phi-3.5-mini-instruct), holding the retrieval task fixed. INT4 quantization significantly reduces accuracy under high interference in every model (e.g., from 81.0% to 68.3% for Qwen), confirmed by paired McNemar's tests ($p \le 2.6 \times 10^{-6}$) and a mixed-effects regression spanning all interference levels; INT8, often assumed safe, also carries a smaller but real penalty in two of three models. The effect is specific to semantically similar (word-type) distractors and reverses sign under a numeric control condition, and is mechanistically linked to a rise in same-key intrusion errors under INT4 (from 21.5% to 24.6% of trials, $p = 4.8 \times 10^{-7}$). A follow-up ablation shows the effect originates in the quantized transformer backbone rather than the output projection layer. These results suggest that bitsandbytes 4-bit quantization can impose an additional cost on applications relying on long, updatable, semantically dense contexts, even when aggregate benchmark accuracy appears largely unaffected. We release our code and tokenizer-verified vocabulary construction method at https://github.com/ShayanShahrabi/compress-and-forget
Yashar Talebirad, Eden Redman, Ali Parsaee +1cs.AI cs.CL cs.IT cs.MA
A bounded agent may obtain information for a decision from its own past, from peers, or from both sources. Retaining task-relevant history can reduce later communication, while a peer message can supply what memory lacks. Under limits on both resources, how should an agent allocate its information budget? Given a fixed task and decision rule, the memory and message rate pairs attaining a performance threshold form an achievable region under specified rules for using history and peer observations. We call its efficient boundary the remembering--signaling frontier. Across conditions where history permits the same maximum reduction in task loss, we hypothesize that a bounded agent will need less peer communication when it obtains a larger loss reduction from history. In preliminary referential games, target repetition coincided with shorter successful messages, while predictability from a hidden cyclic rule did not shorten them. Experiments varying memory and message rates can estimate the frontier and test this prediction across cooperative tasks.
Heesang Ann, Hyunjun Choi, Taehyun Hwang +3stat.ML cs.LG
We study generalized linear bandits with memory, an endogenous non-stationary setting in which rewards depend on past actions through a finite memory matrix. Building on prior work for linear models (Clerici et al., 2024), we show that the previously known $\tilde{O}(T^{3/4})$ regret bound stems from a loose analysis, and we provide a sharpened analysis that recovers a $\tilde{O}(\sqrt{T})$ regret rate in the linear case. We then extend this improvement to generalized linear models and propose a block-wise algorithm based on shrunken confidence bounds. Our algorithm achieves a regret bound of $\tilde{O}\left(\sqrt{mT} + d\sqrt{T} + \sqrtκ\, d^{2} m^{1/4} T^{1/4} + κd^{2} \right)$, where $d$ denotes the feature dimension, $m$ the memory length, and $κ$ a curvature parameter of the link function. This attains a $\sqrt{T}$-type rate despite nonlinear rewards and memory effects. To the best of our knowledge, this analysis provides a unified treatment of memory-induced non-stationarity and nonlinear link functions, while ensuring that the leading regret term is independent of the curvature of the link function. We conduct numerical experiments that are consistent with our theoretical findings.
Embodied intelligent virtual agents are expected to operate as persistent, adaptive, and context-aware entities within complex virtual and Metaverse worlds. However, implementing cognitively capable agents in such environments is conceptually and technologically challenging. Among a range of blueprints and development approaches, the Cognitive Embodied Agent Architecture (CEAA) has been developed as an implementation-oriented framework for architecting components of perception, memory, reasoning, planning, and embodied action. Considering the recent advances in edge computing and generative AI language models, this paper explores the use of Small Language Models (SLMs) to support edge-based operation of selected CEAA components, focusing on "Think" and "Memory" as processes central to cognitive orchestration and persistence of virtual agents in interactive virtual worlds. An edge-based virtual agent gateway system was developed and evaluated on an NVIDIA Jetson Orin NX using Qwen2.5 models of different sizes, exploring the system's capability to process service requests and handle memory-driven conversations. A series of simulation experiments evaluated routing accuracy, memory-read performance, and latency, demonstrating an SLM-driven prototype agent system that partially implements selected CEAA processes to support the development of embodied agents whose cognitive "brain" can operate efficiently and contextually for interactive experiences in immersive virtual worlds.
Users of modern platforms repeatedly need summaries of recent dialogue, but the window rarely contains enough context to be interpreted on its own. We formalize this setting as streaming dialogue summarization, where a system must summarize a current window using selective memory from an unbounded history under a fixed budget. We show that the central challenge is not how much history is accessed, but whether memory recovers the evidence that the current window presupposes. We construct a benchmark and evaluation protocol that separately assesses whether memory contains gap-resolving evidence and whether the generated summary reflects it. We propose ReMEMBER, a missing-evidence memory framework that conditions retrieval on unresolved window dependencies and refines retrieved chunks into evidence-dense memory under a fixed budget. Experiments on dialogues with histories up to 160K tokens show that ReMEMBER improves memory recall and gap-resolution completeness over memory construction baselines under the same budget.
Personalized generation systems retrieve user history by request--memory relevance and inject it into the model context. Yet relevant history may concern the wrong preference aspect, duplicate public information, or provide insufficient support. We argue that personal memory should be used only when it adds utility beyond a public-only response. We propose TRACE-Memory, a two-stage framework for selective personalization. Stage 1 queries for user-specific information missing from the request and public context, then retrieves a coverage-oriented candidate pool. Stage 2 admits a compact subset of source-traceable evidence units, or the empty set, according to response-level incremental utility. We progressively train the query-generation and evidence-admission policies through structured SFT initialization, reduced-space stage-wise GRPO warm-up, and nested multi-sample Joint GRPO. Across 4,500 Controlled and Natural tasks from Goodreads, Amazon Reviews, and Reddit, TRACE-Memory consistently outperforms random and lexical memory use, improves over semantic retrieval, remains competitive with frontier-LLM memory pipelines as local generator capacity increases, and conditions evidence admission on public-context sufficiency, supporting selective rather than default personalization.
Guiyu Zhao, Longteng Guo, Yanghong Mei +7cs.RO cs.CV
While Vision-Language-Action (VLA) models have advanced embodied AI, their fundamentally reactive paradigm severely limits performance in partially observable and long-horizon tasks. When restricted to a single wrist-mounted camera, they inevitably suffer from perception forgetting as objects exit the field of view, and temporal task-progress forgetting} during multi-step execution. To overcome these bottlenecks, we propose AtlasVLA, a novel framework that transitions from direct reactive manipulation to proactive reasoning through a persistent world-ego state. AtlasVLA features a dual-memory architecture: a 4D Persistent World State Memory that lifts transient 2D observations into a globally updated, voxel-hashed spatial state to resolve visual blind spots, and an Ego-Working State Memory that tracks historical ego state and task progress. By conditioning a diffusion transformer (DiT) on this joint World-Ego state, AtlasVLA enables robust spatial reasoning. Extensive evaluations across LIBERO, RLBench, and real-world benchmarks demonstrate that AtlasVLA achieves state-of-the-art performance using solely a wrist camera. Remarkably, it decisively outperforms multi-view baselines, yielding absolute success rate improvements of 9.4% on LIBERO-Long and 17.5% in real-world long-horizon tasks.
Frontier language models solve reasoning problems in a single forward pass that would have been research contributions years ago, yet fail at multi-hour tasks: losing track of earlier decisions, declaring half-finished work done, or drifting from goals. We call this the horizon gap and survey 1,547 arXiv papers (2024-2026) collected via systematic seed harvest with a disclosed 26.8% bleed filter, extended by targeted supplementation. We disambiguate three routinely conflated properties: long-horizon (task property: required steps), long-context (model property: token capacity), and long-term memory (system property: persistence across steps/sessions). We organize the corpus into six categories tracking a long-horizon task's lifecycle -- planning, memory, execution, training, evaluation, and foundations/safety -- crossed with an axis capturing where horizons are carried (within-context, within-task-beyond-context, or cross-task-persistent). Across all categories, we find the same pattern: outcome-only signals grow uninformative as horizons lengthen, and the field's response -- whether process reward models, credit assignment, or trajectory-level diagnostics -- manufactures denser step-level signals. We treat critical and diagnostic literature as first-class threads throughout, arguing that segregating critique from method would routinely split single papers across chapters. We close by naming open measurement problems: decomposing model versus harness capability, managing correlated bias in process-level signals used for both training and evaluation, and whether long-horizon reliability admits general predictive theory.
Memory is essential as language agents move from isolated tasks to long-horizon, stateful workflows, yet existing evaluations often reduce it to retrieval or question answering. We introduce ContextWeave, a longitudinal benchmark that evaluates whether recalled experience improves downstream agent performance in realistic office-work streams. ContextWeave reconstructs privacy-preserved, multi-month workflows of 14 participants into 1,005 executable tasks, including 568 core evaluation tasks, with instructions, containerized environments, trajectories, and task-specific rubrics. It measures workspace quality and alignment with participant-specific preferences, complemented by diagnostics of relevance, continuity, solvability, and robustness to misleading recall. Across six memory components under a fixed model, the strongest configuration raises Workspace Score from 68.08 to 78.20 and Preference Score from 41.50 to 70.60. With a fixed memory component, recall improves both outcomes for all five tested base models, although gains vary substantially. Our analysis shows that actionable, experience-rich memory supports workflow continuation and reduces redundant exploration more effectively than compact summaries, while it can also be more susceptible to misleading recall. These findings motivate memory systems that optimize not only retrieval relevance but also reliable use during execution.
Recursive self-improvement requires agents to turn accumulated experience into better future behavior. Personal AI agents offer a concrete setting for studying this capability because they retain preferences, task histories, tool routines, and learned skills across sessions. Yet whether retained experience actually improves them over time has not been systematically tested. We introduce PAST-Bench, a benchmark designed to isolate this question. Each agent runs through ordered sequences of fresh-session tasks under matched conditions that turn retained experience on and off. It spans 26 scenarios and 204 episodes across memory, procedural reuse, information gathering, and update. We report both later-task gains and whether those gains follow the intended save, retrieve, and update pathway. Across seven base models and four agent frameworks, improvement is real but uneven across capabilities. Agents with the same headline gain can differ markedly in whether that gain is supported by evidence of the intended pathway. Guided by these findings, we develop Hermes+, which extends Hermes with five targeted interventions across stages of the agent loop. Hermes+ raises the average gain from retained experience and provides clearer pathway evidence, with its strongest improvement on tasks requiring outdated state to be replaced, although the effect remains capability- and model-dependent. Together, PAST-Bench and Hermes+ provide an evaluation and diagnostic foundation for studying how persistent agents can progress from retaining experience to systematically improving through it. Code: https://github.com/Gen-Verse/PAST-Bench
We introduce \ours{}, a recurrent Transformer architecture with fixed-size memory that generalizes sliding-window attention while remaining parallelizable during training. \ours{} consists of two coupled models: a prefiller $Q$, which leverages full attention\footnote{In practice, we use interleaved full and sliding-window attention for $Q$, as this yields stronger performance. The essential requirement is that $Q$ be more expressive than $P$, with access to the full history.} to produce memory targets $m'_t$, and a decoder $P$, which uses only sliding-window attention and recurrent K/V injection to produce decoder memories $m_t$ for next-token prediction. We train \ours{} with a memory consistency loss that aligns $m_t$ with $m'_t$, allowing inference to use $P$ alone. Empirically, \ours{} improves validation loss and downstream pretraining benchmarks over sliding-window and latent recurrent transformer baselines. Moreover, sharing parameters between $P$ and $Q$ reduces parameter memory while preserving most of the gains.
Anusha Madan Gopal, Aras Pirbadian, Kristofor D. Carlson +2cs.LG cs.AI cs.IR
Retrieval-augmented generation (RAG) imposes a prefill cost proportional to retrieved context length, and -- with Transformer backbones -- a KV-cache that grows with each generated token. State-Space Models (SSMs) avoid the second cost by construction; we eliminate the first, collapsing prefill from $O(L_{context})$ to $O(1)$ per query. We introduce PRECOG (Pre-Computed Context Injection), a retrieval mechanism that exploits a property unique to SSMs: the fixed-size, position-agnostic recurrent hidden state is a complete summary of everything the model has read. PRECOG pre-encodes document corpora offline as SSM hidden states and injects the best-matching state directly at query time, bypassing in-context re-ingestion entirely. The same state-injection mechanism enables SMC (Structured Memory Consolidation): a hierarchical persistent memory with cognitive-domain clustering, an adjustable fidelity-vs-storage dial, and $O(1)$ session initialization, which consolidates short-term episodic states into long-term semantic memory and fuses both with retrieved corpus states at query time. We demonstrate the system on TENNs-LLM, a 1.2B-parameter gated-SSM language model with a 192 KB hidden state. PRECOG matches in-context RAG answer quality, reducing prefill latency from $\sim$27 s to $<$6 ms on edge hardware -- a $\sim$4500$\times$ speedup that crosses the threshold from unusable to interactive. The mechanism is architecturally impossible for Transformer KV-caches, which are position-entangled and grow linearly with context length.
Zhichen Liu, Ruihan Sun, Hengjie Yang +4cs.CL cs.LG
Long-running assistants and agents consume interaction streams that eventually outgrow the context. Existing context retention, summarization, and retrieval preserve access to selected history, but do not provide a persistent state over the full lifecycle when working context changes. We formulate this missing inference capability as \emph{state continuity under context turnover}: carrying computation forward through a fixed-capacity memory state whose lifetime is independent of the active context. We introduce an intrinsic memory method, \textbf{LiveMem}, which augments a pretrained full-attention LLM with a memory state that preserves the historical information over the whole lifecycle while the main attention path retains a bounded KV window. Context turnover and memory state maintaining, memory-oriented post-training, and state-aware serving jointly make this memory state load bearing after its originating tokens are released. Our experiments show that LiveMem achieves leading overall performance among evaluated systems and other intrinsic memory methods. Experiments on LongMemEval show that LiveMem is able to answer the question based on the memory state, even when the supporting evidence has been removed from the current context, and evidence-distance analysis shows that useful information persists beyond the active window. LiveMem thus establishes state continuity as a distinct and complementary abstraction for continual LLM inference.
Large language model (LLM) agents increasingly operate over long interaction histories, where effective reasoning requires identifying and exploiting task-relevant evidence distributed across past observations and actions. However, useful information encoded in previously computed representations is often underutilized during subsequent generation. We propose \textbf{TransMem}, a lightweight inference-time parametric memory module that transforms sparse historical hidden states from a frozen LLM backbone into reusable memory representations. TransMem uses a lightweight gating network to dynamically apply the latent intervention to the current hidden states, without repeatedly encoding the preceding context. To learn transferable memory utilization rather than task-specific knowledge, we introduce evidence-conditioned self-distillation. A memory-augmented student processes the full context and matches the predictive distribution of an evidence-only teacher that shares the same frozen backbone. Experiments on LoCoMo, HotpotQA, and MemoryAgentBench demonstrate consistent improvements across different model architectures and scales. TransMem yields gains of 11.58--29.25 $F_1$ on LoCoMo and 10.20--13.03 $F_1$ on HotpotQA, while improving the average MemoryAgentBench accuracy from 29.54\% to 40.00\%. These results establish sparse historical hidden states as an effective and efficient memory substrate for long-context LLM agents. Our code is available at https://github.com/Haodong-Lei-Ray/TransMem.
Transformer depth is not used uniformly: lower and middle layers build semantic representations, while upper layers increasingly specialize them for prediction. We turn this division of labor into CoMem (Comprehension Memory), which writes each context chunk only through an intermediate layer, retrieves a fixed number of cached residual states, and recomputes the query-conditioned upper layers over the resulting pack. For a fixed retrieval budget, model-side read compute and memory are independent of stored-context length. We evaluate a continued-trained Qwen3-8B base LM under a unified chat-template-free protocol. The backbone is frozen; the flagship trains only a rank-32 self-distillation LoRA on plain PG19, and we report an adapter-free arm separately. CoMem reaches 97.05 on RULER and 38.27 on LoCoMo versus 34.59 for full-context KV-Direct; the dialogue-memory advantage survives conversation-cluster resampling and an independent judge. Results on additional long-context and long-document tasks expose both the benefits of bounded retrieval and its in-window compression tax. Controlled depth sweeps show that deeper caching lowers per-query recomputation but incurs a fidelity loss that self-distillation substantially repairs. In a separate adapter-free efficiency control on an NVIDIA H20 at 128k, CoMem uses 18.26 GB rather than 89.36 GB and achieves a 7.83x prefill speedup. These results show that long-context memory can be organized along the layer axis, not only the token axis.
Long-horizon LLM agents accumulate reasoning traces, actions, and tool observations that can eventually exceed a model's fixed context window. Existing compaction methods address this limitation by discarding, summarizing, or retrieving earlier information, but they may remove task-critical details or fail to recover them reliably. We propose ARC (Addressable Recall Compaction), a context-management framework that separates archival storage from active-context presentation. ARC stores tool observations in an append-only, ID-addressable log and replaces older observations with compact citations when compaction is required. The agent can subsequently use these identifiers to request stored content without re-executing the corresponding tools or depending solely on similarity-based retrieval. We evaluate ARC using Qwen3-8B with a 16k context window and Qwen3-32B with a 32k context window. On the Needle-in-a-Haystack evaluation, ARC achieves an average exact-answer accuracy of 99.40%, compared with 88.12% for the best-performing baseline in our evaluation. ARC also reduces estimated serving time and HBM traffic under our hardware-cost model. On the LongBench-v2 Hard subset, ARC obtains an average accuracy of 29.97%, compared with 28.25% for the best-performing baseline. These results indicate that explicit, address-based recall can improve information retention and serving efficiency relative to the evaluated context-management baselines under the tested settings.
Long-horizon agents increasingly reuse their KV cache as memory: a serving system keeps a subset of cached entries and drops the rest. Eviction and episodic-memory schemes therefore rest on a premise rarely tested directly, that a retained event is still informative once the observations that produced it are gone. We test it by omitting one earlier observation from what is served, across otherwise identical agent histories. Among items sensitive to that observation, the answer overwhelmingly follows the omitted value, though no served span says which value is correct. We call this semantic materialization: a downstream event's cached rows act as an independently servable view of computation whose inputs are gone. It can also be written on purpose. A deliberately phrased, answer-free event raises donor-aligned recovery from 6% to 51% on Qwen3-8B without ever naming the value, whereas passively harvesting natural mentions from long-term dialog yields no detected advantage. What such a row carries is specific and bounded. Compact state survives, larger payloads decay toward chance, and whether a construction writes at all turns on phrasing rather than on meaning alone, so two phrasings the model comprehends equally well can diverge sharply. The result is a memory contract for sparse event-KV serving: what to write, where it lands, and what survives once the source is gone. For anyone who evicts the corollary is that dropping a source event and observing no accuracy loss does not show the source was unnecessary.
Large language models and LLM-based agents are widely used as personal chat assistants, enterprise copilots, and autonomous workflow agents. In all these applications, memory (the ability to retain, access, and reason over information accumulated over long contexts and multiple interactions) plays a crucial role in determining the reliability of any agent. We introduce RECON (Reasoning over Extended Contexts with Obfuscated Narratives), a benchmark for evaluating compositional reasoning over long contexts. RECON spans 24 case files across three domains (criminal, medical, and financial), each ranging from 50k to 100k tokens, and tests agents on six memory intensive tasks: reconstructing multi-hop evidence chains, propagating cascading invalidations, resolving source conflicts, counterfactual reasoning, satisfying temporal constraints, and temporal fact retrieval. Recent memory benchmarks evaluate whether agents can retrieve scattered facts or detect if a fact has changed whereas RECON evaluates what happens after the change, whether agents can trace which downstream conclusions are affected, which survive through independent support, and how alternative timelines would have unfolded. Our evaluation reveals substantial limitations across current architectures: even the strongest non-Oracle system reaches only 22.4% Accuracy, with retrieval and reasoning each surfacing as challenges.
Embodied agents accumulate experience over time. We study how accumulated experience can be formed into persistent memory for future reasoning and action. We formulate Embodied Action Memory (EAM) as the capability to form and use memory over embodied experience, together with the persistent memory state produced by that process. We introduce MEMORA, a framework that instantiates EAM through a formation-consolidation-retrieval lifecycle and a multi-store world-memory architecture. MEMORA organizes experience into participant-specific Environment, Entity, Activity, and Inferred Knowledge stores: online editing revises memory as new evidence arrives, while offline consolidation abstracts repeated experience into reusable routines, habits, and preferences. We evaluate MEMORA with MEMORA-Bench, a 45-hour egocentric-video suite that measures both retrospective memory faithfulness and prospective memory-grounded planning. Across four open-weight answer models, MEMORA achieves the strongest aggregate planning performance among the evaluated memory interfaces, with its largest gains on out-of-distribution planning. On these tasks, MEMORA improves Robot-Grounded Plan score by up to 16.6 percent, suggesting that memory formed and consolidated across experience can support planning for new goals beyond directly observed episodes. A physical-robot demonstration further shows that memory formed solely from human egocentric video can ground high-level robot plans in participant-specific objects and preferences. Project website: https://github.com/yuzihaowashu/MEMORA
Large Language Model (LLM) agents have shown remarkable capabilities in autonomous decision-making by generating sequential trajectories of states, actions, and observations. However, in complex, long-horizon tasks, these agents frequently suffer from compounding errors and struggle to recover from failures. Existing self-correction mechanisms rely on prompt-based reflection, which is inherently brittle, incurs heavy time and API costs due to iterative trial-and-error loops, and produces task-specific memory that may be hard to generalize to new scenarios. To address this, we propose Experience Memory Graph (EMG), a framework that reformulates agent failure recovery as a graph matching problem. At training time, we convert both failed exploration trajectories and successful expert trajectories into directed action decision graphs. By matching these graphs, we extract common subgraphs (successful workflows) and graph edit paths that explicitly indicate how to correct failures (e.g., which actions to add, delete, or relabel under a given observation), and store them in a memory graph with intra-task nodes and cross-task edges. At test time, EMG retrieves relevant insights and guides the agent in a single, loop-free execution. Experiments on ALFWorld and ScienceWorld show that EMG consistently outperforms state-of-the-art reflection baselines in success rate and average reward, while requiring no test-time trial-and-error.
Gleb Kuzmin, Ivan Rodkin, Aydar Bulatov +8cs.CL cs.AI
Extending the context length of large language models (LLMs) is critical for many real-world applications, yet standard transformers remain constrained by quadratic compute and linear memory scaling. In this work, we investigate the Associative Recurrent Memory Transformer (ARMT) as a practical approach for enabling long-context processing in LLMs, constant memory scaling, and better efficiency. We make three main contributions. First, we construct two domain-specific long-context datasets designed to evaluate realistic workloads, focusing on narrow-domain fine-tuning scenarios. Second, we propose a comprehensive training recipe for ARMT-based context extension, combining continued pre-training, synthetic long-context data generation, curriculum learning, and selective integration of associative memory into chosen model layers. Third, we present an extensive experimental study demonstrating that ARMT-augmented models: (i) process inputs well beyond their original context limits without degrading performance relative to in-limit baselines; (ii) generalize more effectively to out-of-distribution context lengths; and (iii) need 30% less FLOPs while preserving baseline performance within the original context window.
Large language model (LLM) agents increasingly rely on external tools served by shared providers and accessed by heterogeneous downstream agents. Existing approaches improve tool use on the agent side through parameter updates, prompt refinement, or agent-side memory, making tool knowledge difficult to share and limited to behaviors observed in past tasks. We argue that reusable tool knowledge should instead be maintained by the tool provider. We introduce ToolAtlas, a graph-based framework that builds a persistent provider-side tool memory of tool capabilities, failure boundaries, and cross-tool compositions through execution-verified probing. At inference time, agents query the tool memory via adaptive graph traversal. Across two MCP-based benchmarks spanning eight services, ToolAtlas outperforms existing tool-side optimization and agent-side memory baselines by up to 21.61% in pass@1 and 18.61% in pass@4. The same tool memory also transfers across environment instances and agent frameworks without retraining or task-time exploration, yielding up to 24.16%/16.22% and 17.49%/14.27% relative gains in pass@1/pass@4, respectively. Ablation studies show that these gains arise from combining tool-centered memory organization with capability-guided execution probing. These results establish provider-side tool memory as an effective and reusable paradigm for tool servers. Our code is in: https://github.com/PuppyKnightUniversity/ToolAtlas.