Peak Memory

Follow this topic Ask about this topic

Momentum

54 papers in the last four weeks, up 157% on the four weeks before. 0.5% of all new papers.

Jul 13Week of Sep 28

Latest papers 279

Jul 15, 2026cs.AI

Experience Memory Graph: One-Shot Error Correction for Agents

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.

Wenjun Wang, Yuchen Fang, Fengrui Liu +2

University of Electronic Science and Technology of China

Graph-Structured Memories Iterative Self-Correction Error Correction Peak Memory

Jul 15, 2026cs.CL

Memory as a Controlled Process: Learned Adaptive Memory Management for LLM Agents

Large Language Model (LLM) agents increasingly rely on external memory systems to accumulate experience across tasks. Yet nearly all existing approaches, from graph-structured memories to reflective insight stores, access memory through fixed, hand-designed heuristics. We argue that this static view of memory is a core bottleneck for agentic learning because optimal memory behavior is fundamentally context-dependent. The early stages of the tasks, benefit from minimal retrieval because memory is sparse; recurring goal types benefit from plan reuse rather than generic nearest-neighbor lookup; stuck agents benefit from re-retrieval with alternative queries; and across long task streams, the memory store itself must be consolidated and pruned to remain useful. We present Memory as a Controlled Process (MemCon), a framework that models memory operations as a Markov Decision Process and learns an online policy that adaptively decides when, what, and how much to retrieve, when to inject a distilled plan, and when to consolidate or forget. MemCon is backend-agnostic: it wraps any existing memory implementation, learns from task-by-task binary feedback with no pretraining and no additional LLM calls, and uses a lightweight tabular contextual bandit with UCB exploration that converges within tens of tasks. Across 6 benchmarks, 3 agent frameworks, and 3 LLM backbones, MemCon consistently outperforms multiple memory baselines by up to 15.2 points in task success while reducing token consumption by 5--20%.

Eric Hanchen Jiang, Zhi Zhang, Yuchen Wu +11

University of California Los Angeles · University of Washington · Northwestern University

Large Language Model Agents Graph-Structured Memories Peak Memory Agentic Learning

Jul 14, 2026cs.AI

Oracle Agent Memory as an Enterprise Memory Substrate for Long-Horizon AI Agents

Agent memory is a systems problem for long-horizon agents. Practical deployments require retention of task state across extended conversations, recovery of user-specific facts and preferences across sessions, and accumulation of procedural knowledge from prior outcomes. These requirements extend beyond document retrieval: a memory layer must determine which interactions become durable state, how that state is scoped, how it is retrieved under latency constraints, and how it is revised or removed over time. This report studies Oracle Agent Memory as a database-native memory substrate built on Oracle Database. Three themes organize the discussion: memory as a lifecycle spanning ingestion, extraction, consolidation, retrieval, summarization, and revision or removal; a layered architecture that separates an active memory core from a passive memory-store interface with explicit scope control across users, agents, and threads; and evaluation methodology in which downstream task accuracy is complemented by memory-centric measures such as evidence retrieval, recall, latency, and estimated token use. The report summarizes LongMemEval results, reaching 93.8% accuracy, compares Oracle Agent Memory against flat-history baselines, using about 10.7x fewer tokens, and published or reported external baselines where available, and closes with implementation-oriented appendix material covering setup, thread lifecycle, and search semantics.

Richmond Alake, Cesare Bernardis, Paul Cayet +10

Oracle

Long-Term Agent Memory Long-Horizon Agents Peak Memory Oracles Artificial Intelligence Agents State-Tracking

Jul 14, 2026eess.IV

Improving Autonomous Nano-drones Performance via Automated End-to-End Optimization and Deployment of DNNs

The evolution of energy-efficient ultra-low-power (ULP) parallel processors and the diffusion of convolutional neural networks (CNNs) are fueling the advent of autonomous driving nano-sized unmanned aerial vehicles (UAVs). These sub-10 cm robotic platforms are envisioned as next-generation ubiquitous smart-sensors and unobtrusive robotic-helpers. However, the limited computational/memory resources available aboard nano-UAVs introduce the challenge of minimizing and optimizing vision-based CNNs -- which to date require error-prone, labor-intensive iterative development flows. This work explores methodologies and software tools to streamline and automate all the deployment of vision-based CNN navigation on a ULP multicore system-on-chip acting as a mission computer on a Crazyflie 2.1 nano-UAV. We focus on the deployment of PULP-Dronet, a state-of-the-art CNN for autonomous navigation of nano-UAVs, from the initial training to the final closed-loop evaluation. Compared to the original hand-crafted CNN, our results show a 2x reduction of memory footprint and a speedup of 1.6x in inference time while guaranteeing the same prediction accuracy and significantly improving the behavior in the field, achieving: i) obstacle avoidance with a peak braking-speed of 1.65 m/s and improving the speed/braking-space ratio of the baseline, ii) free flight in a familiar environment up to 1.96 m/s (0.5 m/s for the baseline), and iii) lane following on a path featuring a 90 deg turn -- all while using for computation less than 1.6% of the drone's power budget. To foster new applications and future research, we open-source all the software design in a ready-to-run project compatible with the Crazyflie 2.1

Vlad Niculescu, Lorenzo Lamberti, Francesco Conti +2

Integrated Systems Laboratory - ETH Zürich, Switzerland · Department of Electrical, Electronic and Information Engineering - University of Bologna, Italy · Dalle Molle Institute for Artificial Intelligence - USI-SUPSI, Switzerland

Unmanned Aerial Vehicles Autonomous Navigation Nvidia Jetson Orin Nano Peak Memory

Jul 13, 2026cs.LG

ToolAtlas: Learning Once, Reusing Everywhere with Tool-Side Memory

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.

Yue Fang, Zhibang Yang, Fangkai Yang +5

School of Computer Science, Peking University, Beijing, China · Microsoft

Reusable Knowledge Peak Memory Atlas

Jul 12, 2026cs.LG

Infrared Organization and Critical Cognitive Field Formation in Transformer Dynamics

Large language models exhibit remarkable emergent behaviors, yet the physical mechanism governing their collective dynamics remains poorly understood. Cognitive Field Theory predicts that learning reorganizes the time-scale density of states (TDOS) through the infrared accumulation of slow relaxation modes, thereby enhancing the memory self-energy, reducing the cognitive forgetting gap, and strengthening the collective susceptibility. Using publicly available Pythia language models, we extract relaxation spectra directly from Transformer layer Jacobians throughout training, network depth, and model scale, allowing the TDOS, memory self-energy, forgetting gap, memory kernel, and infrared critical exponent to be measured quantitatively. The measurements reveal progressive infrared accumulation of slow relaxation modes, producing an approximately flat infrared TDOS with

ρ(λ)\simλ^{-0.1}

and scale-free memory kernels

K(t)\sim t^{-1}.

The memory self-energy exhibits a pronounced transient maximum during early optimization before relaxing toward a metastable near-critical regime, corresponding to the smallest cognitive forgetting gap and the largest collective susceptibility predicted by Cognitive Field Theory. These observations provide quantitative experimental evidence that Transformer dynamics are governed by infrared collective organization. The reproducibility of the same dynamical behavior across training, network depth, and model scales suggests that infrared slow-mode organization represents a universal collective principle of Transformer dynamics.

Byung Gyu Chae

Electronics and Telecommunications Research Institute, 218 Gajeong-ro, Yuseong-gu, Daejeon 34129, Republic of Korea

Slow Relaxation Modes Cognitive Science Peak Memory Infrared

Jul 12, 2026cs.LG

LayerNorm as Implicit Gain Control in Looped Transformers

In pre-LayerNorm looped transformers, LayerNorm inside the recurrent block acts as an implicit gain controller: by coupling the block's local Lipschitz constant inversely to the activation scale, it renders the recurrence Jacobian non-normal -- asymptotically contractive at every verified fixed point even where its operator norm exceeds 1 -- so the true stability budget is the spectral margin, not an operator-norm bound. That margin depletes as the carry

ρ\to 1

, and a minority of initializations never converge to a fixed point at all, so the diagonal carry constraint

ρ(\bar{A}) < 1

is necessary but not sufficient for convergence of the full recurrence. Training experiments across six tasks, including a controlled ablation, reveal that the linear carry is not the depth-memory mechanism: gradient descent routes memory through the block's more expressive nonlinear recurrence and leaves the stability-constrained carry at rest -- the carry's role is stabilization, not memory. We characterize the boundary of this claim: on tasks with axis-aligned per-channel structure, gradient descent does recruit the carry. All results are derived analytically and verified in a from-scratch, CPU-scale implementation; verification at larger scale is needed.

Matthias M. M. Buehlmaier

Faculty of Business and Economics, The University of Hong Kong · Pokfulam Road, Hong Kong

Batch Normalization Recurrent Model Transformer Architectures Gradient Descent Peak Memory

Jul 11, 2026cs.LG

PhysMRV: Physical Memory Retrieval and Verification for Physics Plausibility Reasoning

Video-language models (VLMs) have achieved remarkable performance on video understanding and visual question answering, yet they remain unreliable in reasoning about physical plausibility, where understanding object interactions, causal dynamics, and fundamental physical principles is essential. This limitation is particularly evident on challenging physical reasoning benchmarks, revealing a persistent gap in physical commonsense reasoning. To address this challenge, we propose PhysMRV, a training-free physical memory and verification framework for physical plausibility reasoning. Unlike retrieval-augmented VLMs that retrieve semantically similar videos as additional context, PhysMRV transforms training videos into a Hierarchical Memory Bank of structured physical knowledge comprising three complementary levels: scene descriptions capturing visual context, physical-event graphs modeling object interactions and causal structure, and physics-rule summaries distilling reusable physical principles and cues. During inference, PhysMRV retrieves physically relevant memories and leverages their structured physical evidence to guide a frozen VLM in verifying physical plausibility, requiring neither fine-tuning nor parameter updates. We evaluate PhysMRV on three challenging physical reasoning benchmarks, ImplausiBench, IntPhys2, and GRASP Level 2, across multiple state-of-the-art VLMs. Experimental results demonstrate consistent improvements over direct prompting across diverse VLMs and evaluation benchmarks, showing that structured physical memories provide an effective and scalable means of enhancing physical plausibility reasoning without additional training.

Wenyuan Wang, Lianyu Hu, Hao Wang +1

1Nanyang Technological University · 2Rutgers University · University of Illinois Urbana-Champaign

Physical Plausibility Video-Language Models Peak Memory

Jul 10, 2026cs.AI

A Symbolic Neural CPU for Quantization-Simulated Writeback and Interpretable Program Execution

Neural networks can learn algorithmic input-output mappings, but trusting a learned executor requires more than a correct final answer because the state transitions that produce it are usually hidden. To make those transitions visible, we introduce a trace-supervised symbolic neural CPU, a factorized learned execution architecture that combines recurrent control, an explicit operation router over a fixed differentiable arithmetic-logic unit bank, destination-masked register writeback, complete trajectory supervision and matched fixed-point replay. The model exposes the selected operation, source and destination registers, register trajectory, memory signals and writeback semantics at every step. On the principal 16-wide benchmark, the non-quantized executor reproduces reference execution exactly, while the eight-bit quantization-simulated executor preserves the symbolic operation path through programs of 1,000 instructions. When the same execution is evaluated against a matched fixed-point replay, the residual numerical drift disappears, showing that it comes from a mismatch between continuous and low-precision reference semantics rather than from execution failure. We compare recurrent, Transformer, temporal-convolution, temporal graph-inspired and state-space controllers, and the ablations show that operation-gate supervision is necessary for an inspectable execution path. Hidden-opcode memory-pressure tasks expose the remaining limits in delayed state use and temporal binding. We also extend the interface with ValueMemory, hybrid adaptive leaky integrate-and-fire controllers, candidate-constrained symbolic control trained through behaviour cloning and actor-critic reinforcement learning, and an RV32I base-integer semantic bridge. Together, these results establish a trace-verifiable framework for interpretable, low-precision and controllable neural execution.

Jose Luis Lima de Jesus Silva

Federal University of Bahia, Department of Geophysics, Salvador, BA 40170-115, Brazil · Grupo de Estudos e Aplicação de Inteligência Artificial em Geofísica (GAIA), Federal University of Bahia, Salvador, BA 40170-115, Brazil

Neuro-Symbolic Framework Quantized Self-Speculative Peak Memory

Jul 9, 2026cs.SD

A Quantized Native Runtime for On-Device Semantic Audio Generation

Semantic audio applications increasingly require controllable generation on commodity and embedded hardware rather than through framework-heavy datacenter stacks. We present \textit{aria}, a dependency-free native runtime that runs the complete text-to-music pipeline of Stable Audio~~3 (SA3) on ordinary GPUs, CPU-only machines, and a Raspberry~~Pi~5, with no Python or deep-learning framework underneath. Our main contribution is a study of quantization: running the model at lower numerical precision to fit tight memory budgets, saving memory in place rather than adding to it. Because the runtime owns every internal tensor, it also exposes activation steering, a low-cost way to steer what the model generates. We judge the quality cost with three independent measures of the output (prompt adherence, overall audio quality, taste preservation), each compared against the ordinary variation between random seeds. Eight-bit precision shows no measurable quality loss on any measure while sharply cutting memory, and it is the fastest mode on the GPU; four-bit adds a small, bounded cost but shrinks the footprint enough to run the

1.2

-billion-parameter model on an

8

,GB Pi. Against the official implementation, aria matches or exceeds generation speed and starts about seven times faster. A case study of the steering interface generates music carrying taste associations (\emph{sonic seasoning}), with genuine but bounded control for a subset of attributes. These results make a compact, quantized runtime with built-in control a practical basis for on-device semantic audio in Internet-of-Sounds settings. The \textit{aria} runtime is released at https://github.com/matteospanio/aria.

Matteo Spanio, Antonio Rodà

Centro di Sonologia Computazionale (CSC) · Dept. of Information Engineering, University of Padova

Jul 8, 2026cs.SE

The Memory Wall of Green Software: Empirical Energy Evaluation of Memento Design Pattern

As Green Software Engineering matures, energy efficiency has transitioned into a mission-critical non-functional requirement. While software design patterns ensure structural integrity, their inherent abstraction layers impose an implicit "metabolic cost" that often remains obscured during the design phase. This paper empirically investigates the energy dynamics of the Memento design pattern, contrasting a direct, unabstracted baseline against Classic full-snapshot and Differential delta-encoding strategies. Leveraging the RAPL interface for high-fidelity hardware telemetry, we quantify energy dissipation across state volumes scaling from 10 MB to 200 MB. Our empirical results expose a critical architectural trade-off: the Differential strategy minimizes memory traffic, yielding a maximum energy reduction of 65.8% for mid-scale states, but collides with a catastrophic "memory wall" at 200 MB. At this saturation point, algorithmic optimizations are completely neutralized by severe GC thrashing and non-linear power spikes. We synthesize these findings into evidence-based heuristics, providing architects with a robust framework to reconcile structural design quality with sustainable Green IT imperatives.

Imane Jriri, Tarik Houichime, Younes El Amrani

LRIT, Faculty of Science, Mohammed V University In Rabat, Rabat, 10112, Morocco · Meridian Team, LyRICA Laboratory, School of Information Sciences, Rabat 10100, Morocco

Peak Memory Software Engineering Empirical Study Software Network Traffic

Jul 8, 2026cs.DB

Breaking Database Lock-in: Agentic Regeneration of High Performance Storage Readers for Database Bypass

Analytical workloads operating on data stored in external database systems face a fundamental bottleneck: data access is guarded entirely by the database driver, like JDBC or ODBC, forcing all reads through query execution and other driver layers that are not designed for bulk columnar analytics. We present Jailbreak, an approach that bypasses the database engine entirely by reading storage files directly and materializing data as in-memory columnar buffers. Jailbreak's key insight is that database file formats, while complex, are fully specified by their source code and documentation, artifacts that Large Language Models (LLMs) can ingest to regenerate operator-specific table reading components without human-engineered parsing logic. Jailbreak leverages LLM-assisted code synthesis for database storage decoding, turning a traditionally opaque format into a directly queryable artifact. We evaluate Jailbreak on PostgreSQL and MySQL storage files, targeting analytical snapshot scenarios common in read replicas and offline processing pipelines. The generated reader produces Apache Arrow buffers consumable directly by most of the widely known query engines, including DuckDB, Apache Spark, and GPU-accelerated frameworks such as cuDF and Spark RAPIDS. We validate correctness against JDBC/ODBC-based baselines using the TPC-H benchmark across all query results, and demonstrate significant performance improvements in end-to-end analytical throughput, achieving up to 27x speedups. Our results showcase that LLM-assisted storage reader synthesis is a viable and generalizable methodology for breaking data lock-in across database systems, with applications beyond PostgreSQL and MySQL for any system whose file format is available to the LLM from documentation or source code.

Victor Giannakouris, Immanuel Trummer

Cornell University Ithaca, NY, USA

Peak Memory Jailbreak Success Rates

Jul 8, 2026econ.GN

Memory Scarcity, Open Models, and the Restructuring of the AI Industry, 2026-2030 -- A quantitative scenario analysis of inference economics, training-cost divergence, and infrastructure solvency

We analyze how four forces restructure the AI industry over 2026-2030: the DRAM/HBM price surge, frontier-capable open-weight models (GLM-5.2), rapid inference-efficiency gains (near-Shannon-limit KV-cache compression, lightweight local runtimes), and the entry of Meta and xAI into compute resale on fleets bought before the memory repricing. Formulating inference economics in dollars per petabyte of bandwidth delivered ($/PB) -- model-agnostic for bandwidth-bound decode -- we show the entrant-incumbent cost gap never closes: a depreciation conveyor delivers newly amortized fleets to incumbents faster than hardware prices normalize (3.2x in 2026, 1.9x in 2027, re-widening to 3-4x by 2029-30). Training bifurcates into a luxury tier ($18-38B per frontier run by 2030) and a mass tier (previous-frontier parity via RL/distillation falling toward $5M). Solvency of the announced buildout is confined to a corridor requiring roughly 2x annual token-demand growth for four years with sticky premium pricing; a measurement critique shows public token trackers overstate monetizable demand, and all pre-Q2-2026 projections predate the industry's shift from token maximization to token minimization. A vintage-breakeven analysis finds 2026 and 2028-29 capacity each fatally exposed to one pricing regime, with only the 2027 vintage robust. A greenfield custom-silicon entrant removes the merchant margin but not the memory premium (central outcome: 25% success/34% mediocre/41% loss, improvable via staged go/no-go gates). China's LineShine LX2 -- domestic HBM on a standard ISA -- decouples its cost curve from the memory crisis. Scenario probabilities: Rotating Landlord Oligopoly 25%, Commoditization Crash 25%, Jevons Absorption 20%, System-Layer Re-differentiation 18%, Geopolitical Bifurcation 12%. Solvency now depends on monetized bandwidth demand, premium stickiness, and vintage ownership.

Satoshi Matsuoka

RIKEN Center for Computational Science (R-CCS)

Artificial Intelligence Infrastructure Economies Peak Memory

Jul 8, 2026cs.LG

Fractal KV-Cache Archives: Lossless Symbolic Storage with In-Place Retrieval for Long-Context LLM Inference

The key-value (KV) cache dominates the memory cost of long-context autoregressive inference, and a growing body of work compresses it through quantization, eviction, or offloading. We study a complementary question: once a position's KV state has been quantized to codebook indices, how should the resulting symbol stream be stored, and can the storage layer do more than store? A family of contractive iterated-map codes that serialize a symbol sequence into a sequence of low-dimensional real vectors is revisited, and it is shown that they form a natural archive format for a quantized KV cache with the following features. The method provides exactly the access pattern a growing cache requires. It is lossless, it runs in linear time, and supports O(1) random access and O(1) amortized append. A controlled study of the quantizer feeding this archive is conducted on GPT-2 with 1024-token contexts. Keeping a small exact window (4 attention sinks + 32 recent tokens) and archiving the rest, per-head residual vector quantization reduces the archived cache by 36-54x relative to an fp16 cache at a perplexity cost of 11-15%, and we quantify a sharp key/value asymmetry -- quantizing keys is roughly 4x more damaging than quantizing values, consistent with prior low-bit KV work -- and use it to allocate bits in a hybrid scheme. Finally, we show the archive is simultaneously a search index: approximate substring queries execute directly on the stored vectors, and matched context is decoded from the matched vector without ever materializing the surrounding text. We release all code; every number reproduces from a single command on a laptop CPU.

Vladimir Gusev

Independent Researcher

Key-Value Cache LLM Inference Optimization Quantizer Peak Memory Autoregressive Model

Jul 7, 2026cs.AI

Akashic: A Low-Overhead LLM Inference Service with MemAttention

Recent LLM-based agent systems continuously accumulate context across multi-turn interactions, tool invocations, and cross-session workflows. Replaying the full history for every request quickly becomes impractical: long contexts increase prefill cost, may exceed context limits, and often bury task-relevant evidence in irrelevant content, degrading both serving efficiency and output quality. We propose Akashic, a low-overhead memory system built around MemAttention, which organizes context into bounded chunks and models semantic relationships across chunks, preserving cross-chunk evidence without repeatedly rewriting the full history. Akashic further applies hardware-software co-designed memory placement to co-locate likely co-retrieved chunks, reducing retrieval fragmentation and I/O overhead. Across four representative workloads and three model sizes, Akashic improves task accuracy by up to 10.2 points, throughput by up to 1.21x, and sustainable request rate by up to 1.88x over strong prior memory baselines.

Yang Liu, Zhaokai Luo, Huayi Jin +7

Xiaohongshu Inc., China · ShangHai JiaoTong University · Peking University

Large Language Model Memory Peak Memory LLM Inference Optimization Linear Attention Fragmentation

Jul 7, 2026cs.CV

MiLSD: A Micro Line-Segment Detector for Resource-Constrained Devices

Line segment detection is a key building block in visual SLAM, 3D reconstruction, and industrial inspection. Recent deep learning methods have greatly improved accuracy, yet even the smallest models require several megabytes of memory, exceeding low-cost MCU capacity. This work investigates the maximum achievable accuracy under a sub-megabyte budget. We propose MiLSD, a detector tailored for MCU-level constraints, and systematically compare three output representations within a compact fully-convolutional backbone. Our study shows that the proposed F-Clip center-with-length-and-angle formulation learns most effectively at small model sizes. We find that 8-bit quantization preserves full-precision performance, while 4-bit quantization causes significant degradation, particularly in angle regression, with quantization-aware training recovering only part of the loss. With a one-megabyte activation budget and inference enhancements including sub-pixel decoding, test-time augmentation, and a lightweight verifier, MiLSD improves sAP10 on ShanghaiTech Wireframe from 10.6 (25k parameters, 0.25 MB) to 24.1 within 1 MB. Rather than competing with GPU-scale parsers, we map the accuracy memory trade-off across representations, bit-widths, capacities, and post-processing strategies for embedded vision systems.

Parsa Hassani Shariat Panahi, Amir Hossein Jalilvand, M. Hassan Najafi

Mobilenetv2 Resource-Constrained Edge Devices Sequential Visual Localization Peak Memory

Jul 2, 2026cs.LG

Training Hybrid Block Diffusion Language Models with Partial Bidirectionality

High-throughput long-context generation is one of the central challenges for large language models. Generation is typically memory-bandwidth-bound rather than compute-bound: each decoding step must stream the accumulated key/value (KV) cache from memory, so bandwidth demand grows with context length while only one token is emitted. Two parallel approaches have therefore emerged: reducing memory access with efficient attention variants and linear-time mixers such as Mamba, or increasing parallel computation by generating blocks of tokens at once. However, technical challenges arise when combining these two ideas. Earlier hybrid diffusion models such as DiffuMamba use bidirectional Mamba mixing, including a reverse-direction scan relative to causal generation. This reverse scan needs to scan the entire sequence, so its states are not prefix-only and cannot be precisely reused as a cache even when diffusion is performed block by block. We propose a BDLM Mamba--attention hybrid that addresses this challenge by restricting the reverse Mamba scan to the active denoising block, which enables exact caching across blocks. In an 87M-parameter DCLM sweep, BDLM Mamba-H achieves the best C4-en validation perplexity compared to BDLM attention and full-sequence baselines. At 350M parameters, it remains competitive with BDLM attention. For long-context inference, BDLM Mamba-H reaches 19.7x the throughput of full-sequence DiffuMamba-H at 65K tokens and 3.7x the throughput of BDLM attention at 262K, showing that Mamba hybrids are a potential long-context diffusion architecture.

Pranshu Chaturvedi, Parth Shroff, Tarun Suresh +2

Stanford University

Diffusion Language Models Efficient Long-Context Inference Diffusion Models Bidirectional Topomamba Peak Memory

Jul 2, 2026cs.AI

DRIFTLENS: Measuring Memory-Induced Reasoning Drift in Personalized Language Models

Personalization changes what a model says to a user; we show that it can also change the reasoning trajectory used to justify the response. Modern LLMs personalize interactions by storing user attributes, preferences, and prior context, then injecting this information into future prompts. We study whether such memory reshapes reasoning on open-ended questions where no single ground-truth answer exists. To quantify this effect, we introduce DRIFTLENS, a ground-truth-free framework that maps each expressed reasoning step to a value category and measures divergence between a question's no-memory trajectory and its trajectory under injected user-attribute memory. We first validate that DRIFTLENS distinguishes content-free pragmatic noise from substantive reasoning changes. Across four LLMs and 10 user-attribute categories, including age, occupation, and disability, user-attribute memory induces medium-to-large reasoning drift above each model's pragmatic-noise floor, even when final answers remain fluent, on-topic, and plausible. We then evaluate GRPO- and DPO-based post-training methods for reducing drift. Both reduce drift, but neither uniformly dominates; effects on downstream capability, helpfulness, and instruction following are model-and reward-dependent. These results suggest that memory-induced reasoning drift is a measurable and only partly mitigated failure mode of personalized language models.

Xi Fang, Weijie Xu, Yingqiang Ge +3

1Amazon

Large Language Model Personalization LLM Reasoning Strategies Large Language Model Memory Peak Memory Open-Ended Responses Lens

Jul 2, 2026cs.AI

A Hippocampus for Linear Attention: An Exact Memory for What the Recurrent State Forgets

Linear-attention and state-space language models compress the prefix into a fixed-size recurrent state, yielding O(1) memory at the cost of a lossy exact memory: when many key--value associations compete, earlier facts are overwritten and needle recall degrades. Inspired by Complementary Learning Systems, we give linear attention a hippocampal complement. HOLA (Hippocampal Linear Attention) keeps the usual delta-rule state as a compressive memory and adds a bounded exact KV cache, forming a semiparametric test-time memory: the state models linearly compressible structure, while the cache stores associations that should not be forced through that state. The cache writes without a learned eviction module, keeping tokens with large beta * ||e||, the prediction residual actually committed to the state; a decoupled RMSNorm-gamma cache read then turns these exact KV pairs into sharp retrieval rather than soft averaging. At 340M parameters trained on 15B SlimPajama tokens, HOLA lowers Wikitext perplexity from 27.32 to 22.92 (-16.1%), below a full-attention Transformer++ (26.88), and improves LAMBADA perplexity from 30.95 to 30.26. It also achieves the best linear in-context retrieval and remains much more robust than GDN or a matched HOLA+recency cache on RULER needle-in-a-haystack recall out to 32k tokens (16x its training length).

Wanyun Cui

Shanghai University of Finance and Economics

Kimi Delta Attention Linear Attention Recurrent State Efficient Long-Context Inference Peak Memory In-Context+1

Jul 2, 2026cs.CV

Efficient PEFT Methods with Adaptive Checkpointing for Vision Models and VLMs on Resource Constrained Consumer-GPUs

Modern pretrained vision models achieve strong accuracy but demand substantial GPU memory for fine-tuning, making edge deployment impractical. This paper compares five parameter-efficient fine-tuning (PEFT) methods (Full FT, LoRA, AdaLoRA, QLoRA, BitFit) on Transformers- (ViT-Small, TinyViT) and Mamba-based vision backbones (Vim-Small, MambaVision-T) under an on-device VRAM budget (e.g., 2 GB), together with three gradient-checkpointing strategies (none, static, and a proposed memory-budget-aware adaptive algorithm); and we evaluate three families of foundation-model baselines: zero-shot contrastive vision language models (OpenCLIP, SigLIP), self-supervised vision backbones with lightweight evaluation protocols (DINOv2), and autoregressive VLMs for prompt-based classification (PaliGemma, MobileVLM, SmolVLM). Experiments on CIFAR-100 and DTD report accuracy, training time, energy, and the NetScore family of multi-objective metrics, which we extend with two deployment-aware variants. QLoRA and BitFit cut energy 20-30% at a 1-2% accuracy cost; the adaptive algorithm reduces peak memory 43-79% with 9-30% energy overhead. DINOv2 surpasses fine-tuned models on CIFAR-100 (0.917 vs. 0.897) at a fraction of the energy, while small autoregressive VLMs remain uncompetitive.

Altay Toktassyn, Jurn-Gyu Park

Department of Computer Science, Nazarbayev University, Astana 010000, Kazakhstan

Parameter-Efficient Fine-Tuning Methods Vision Foundation Models Cpu-Gpu Hybrid Designs Vram Model Checkpoints Peak Memory+1

Jun 30, 2026cs.AI

From Signals to Structure: How Memory Architecture Drives Language Emergence in LLM Agents

How do two agents invent a shared language from scratch? In a Lewis signaling game, a sender and receiver must coordinate on a code using only their interaction history. We study five memory architectures across varying channel configurations with LLM agents and find that memory architecture matters more than channel capacity. Agents with a persistent private notebook benefit from surplus channel capacity and avoid the high-capacity collapse seen in stateless agents, achieving the most reliable coordination (

0.867 \pm 0.023

at capacity = 25). Stateless agents peak at moderate capacity and then degrade as the vocabulary grows beyond what a rolling context window can track The notebook externalizes learned conventions, freeing agents from having to re-derive codes each round. An information bottleneck-inspired argument predicts an optimal capacity equal to the number of objects. Instead, the bottleneck (capacity = 8) proves to be a fragility point, and surplus capacity is generally better. We show that channel capacity alone cannot predict coordination; memory architecture determines whether agents turn interaction history into stable conventions, and both dimensions are needed to understand how signals become language.

Yashar Talebirad, Eden Redman, Ali Parsaee +1

Alberta Machine Intelligence Institute, University of Alberta, Edmonton, Canada · Network for Applied Technology, Edmonton, Canada

Large Language Model Agents Multi-Agent Coordination Interaction History Peak Memory Bottlenecks Channel Modeling

Jun 30, 2026cs.CV

PriorEye: Geospatial Visual Priors for End-to-End Autonomous Driving

Most end-to-end autonomous driving methods rely solely on instantaneous sensor observations, limiting them to reactive behavior without the anticipatory foresight human drivers employ through prior experience. We introduce geospatial visual priors, street-level visual context anchored to the intended driving route, providing visual-spatial foresight independent of real-time sensors. We propose a memory augmentation module featuring a dual-memory architecture and an adaptive memory gate, which can be easily integrated into existing end-to-end approaches. This design pairs a contextual memory for retrieved priors with a persistent fallback memory, and dynamically regulates the influence of memories based on current state compatibility. Evaluated on the NAVSIM-v2 benchmark, our approach consistently improves performance across diverse end-to-end baselines. Furthermore, because these priors are independent of onboard sensors, our method inherently improves robustness against sensor corruption, while the dual-memory design ensures safe fallback when the retrieved priors themselves become unreliable. Our project page is available at https://ori-mrg.github.io/PriorEye.

Kyuhwan Yeon, Benjamin Ramtoula, Daniele De Martini

Mobile Robotics Group, University of Oxford, United Kingdom

Autonomous Driving Spatial Priors Autonomous Navigation Anticipate Peak Memory

Jun 30, 2026cs.LG

ECHO: Prune To Act, Trace To Learn With Selective Turn Memory In Agentic RL

Long-horizon language agents must repeatedly interact with tools, accumulate evidence, and make decisions under bounded context windows. Context-management methods make such rollouts feasible by simplifying past interactions through deletion, folding, or memory editing. However, when useful history is collapsed into compressed states, the reconstructed context may no longer reveal which earlier observations support a successful final answer. This creates a mismatch between bounded-context acting and outcome-based reinforcement learning: the policy acts on reconstructed context, while the learner lacks source-level provenance for assigning credit to the evidence that mattered. We propose ECHO, a selective turn-memory framework for traceable context reconstruction in Agentic RL. ECHO compresses each completed environment turn into a compact source-indexed memory record, reconstructs bounded policy contexts by selecting useful records, and reuses the selected source indices to route positive outcome credit to the final trajectory segment, reused evidence turns, memory findings, and memory-selection actions. On BrowseComp-Plus, ECHO reaches 43.4% held-out accuracy, outperforming GRPO at 28.9% and the rolling-summary baseline SUPO at 36.1%, while using fewer turns and lower trajectory volume than SUPO. The trained policy also improves zero-shot generalization across multi-objective QA, code generation, and deep information-seeking benchmarks on both dense and MoE backbones.

Zijun Xie, Binbin Zheng, Enlei Gong +7

School of Mathematical Sciences, Peking University · Baidu Inc. · University of Science and Technology of China

Agentic Reinforcement Learning Traces Peak Memory

Jun 29, 2026cs.LG

Neural Subspace Reallocation: Continual Learning as Retrieval-Based Subspace Memory Management

We introduce Neural Subspace Reallocation (NSR), which reframes continual learning as memory management over parameter subspaces. Instead of treating Low-Rank Adaptation (LoRA) modules as disposable per-task adapters, NSR manages them as compressible, retrievable memory units on a frozen backbone through a recurring cycle: (1) compress learned LoRAs via SVD, (2) reserve them in a TaskKnowledgeBank, (3) recall related past LoRAs by embedding similarity to warm-start new or returning tasks, and (4) reallocate the active subspace accordingly, with distillation protecting prior tasks. We prove that in cyclic environments any memoryless allocation policy incurs cumulative regret Omega(T(M-1)Delta_switch) relative to a history-aware policy backed by the Bank (Theorem 1). Empirically, on Split-CIFAR-100 the Bank reduces cyclic recovery time by 10x, exactly as predicted, and on the heterogeneous 5-Datasets benchmark NSR achieves the highest accuracy and the least forgetting, about 9x closer to zero backward transfer than the memoryless heuristics. Crucially, we run a controlled study that isolates which component matters: holding the Bank fixed and varying only the allocation rule, we find that a simple similarity-based retrieval rule matches or beats a learned reinforcement-learning controller (recovering recurring tasks in 0 vs 1.8 steps and reaching equal accuracy). Our central, honest finding is therefore that the memory mechanism -- compression and similarity retrieval -- rather than a learned allocation policy, drives continual-learning performance under fixed capacity. A memory-budget analysis confirms the compressed Bank stays small -- 0.29 MB of parameter memory per task -- so a top-K retention cap bounds the total footprint while preserving fast recovery for retained tasks.

Byeong Hoon Yoon

Independent Researcher

Replay-Based Continual Learning Continual Learning Subspace Peak Memory Forgetting

Jun 28, 2026cs.LG

Optimizer Memory Makes Shuffle Order a First-Order Source of Fine-Tuning Noise

Shuffle order can be a larger source of fine-tuning noise than a memoryless analysis predicts: fixed-clock optimizer memory makes local equal-multiset contrasts first order in the learning rate rather than second order, and the resulting order channel can be large enough for a single seed to flip a close A/B comparison. We isolate this mechanism and derive a fit-free way to size the noise it produces. For a memoryless optimizer, reordering an equal multiset has no first-order endpoint term; the leading local contrast is the

O(η^2)

gradient bracket. Fixed-clock optimizers such as AdamW are different. Their moment buffers, preconditioner state, and de-biasing counters advance with the step index rather than with the learning-rate-scaled time

τ=ηk

, so the same gradient can receive a position-dependent endpoint weight. For any fixed finite measurement window, a lifted-state expansion gives an

O(η)

equal-multiset contrast whenever the first-order replay coefficient is nonzero, while regular and clock-matched controls remain

O(η^2)

; a bare fixed-

β

momentum buffer is already enough. A bitwise-deterministic replay from one warmed optimizer state isolates the mechanism, giving order-variance slopes 1.83 for AdamW, 2.00 for fixed-

β

momentum, and 4.00 for SGD; matching the memory clock to

τ

restores the regular exponent. For AdamW with a frozen preconditioner, the same impulse-weight kernel gives a closed-form asymptotic order-variance floor after the local potentials are measured, with no fitted coefficients. The result is local to the measurement window (independent reshuffling can average the channel across windows), but it yields order-noise error bars, positional attribution weights, and a seed-budget criterion for fine-tuning comparisons.

John Sweeney

Random Rotations Prioritized Experience Replay Finite Time Analysis Peak Memory Model Fine-Tuning Contrastive Learning

Jun 27, 2026cs.CL

MMLA: Memory-Mediated Learning Architecture for Predictive Dual-State Adaptation

Memory-Mediated Learning Architecture (MMLA) separates slow base parameters theta, a bounded numerical policy carrier Phi, and a bounded authoritative memory M. Predictive Dual-State Adaptation (PDSA) lets feedback update Phi while one problem remains active and lets a trusted lifecycle atomically commit one typed row or exact NULL. Later reasoning may read both states, but their writers, resets, rollback domains, and ledgers remain distinct. Realized futures supervise values only during training; deployment is causal and future-blind. We give conditional theory and falsifiable contracts for reasoning-time updates, completed-segment consolidation, predictive admission, authoritative memory, and dual-state attribution. Assumptions, counterexamples, capacity and cost ledgers, recovery duties, and identifying experiments are explicit; these are not implementation guarantees. Controlled studies show exact lifecycle execution on 300/300 held-out records for each of three seeds, calibrated retrieval gains over frozen-hidden dense and BM25 baselines, and exact typed anchor-filler transport on 240/240 held-out records per seed. These validate restricted components, not natural-language memory management or complete PDSA. Post-training studies retain positive and negative evidence. Later protocols obtain restricted readout progress, but a nine-trajectory comparison finds that matched latent readout, a full-width bridge, and bridge plus frozen text-teacher alignment all fail continuous-event qualification across both task families. A separately adapted text reference and restoration checks pass. At the September 13, 2026 evidence cutoff, no strict policy-only reasoning-time-training effect, predictive-admission oracle margin, learned future-blind admission policy, or policy-by-memory factorial advantage is established.

Junyi Zou, Avrova Donz

1MMLA-org · 2Communication University of China (CUC), No. 1 Dingfuzhuang East Street, Chaoyang District, Beijing 100024, China

Efficient Long-Context Inference Long-Term Memory Peak Memory Preliminary Study

Jun 26, 2026cs.SE

SWE-MeM: Learning Adaptive Memory Management for Long-Horizon Coding Agents

Long-horizon software engineering agents often need to manage lengthy and noisy interaction histories under limited context budgets. Existing memory management methods typically rely on static compression workflows or impose rigid constraints on compression timing and granularity. Moreover, these approaches fail to jointly optimize memory management and issue resolution capabilities to improve performance while reducing token usage. We present SWE-MeM, a training framework for proactive and on-demand memory management in software engineering agents. SWE-MeM provides a flexible memory tool that lets agents decide when, what, and how to compress based on trajectory state, task progress, and remaining context budget. We train agents with synthesized proactive memory-management trajectories and Memory-aware GRPO, which jointly optimizes memory management and issue resolution through memory-aware trajectory splitting and step-level credit assignment. On SWE-Bench Verified, SWE-MeM achieves 43.4% and 60.2% resolve rate with 4B and 30B models, respectively, outperforming existing memory management baselines in both performance and efficiency.

Shuzheng Gao, Wenhao Zeng, Zhaojian Yu +5

The Chinese University of Hong Kong, China · Shanghai Jiao Tong University, China · Tsinghua University, China +1

Long-Term Agent Memory Peak Memory Coding Agents Repair-Based Group-Relative Policy Optimization

Jun 25, 2026cs.CL

Cascaded Multi-Granularity Pruning for On-Device LLM Inference in Industrial IoT

Deploying large language models (LLMs) on Industrial Internet of Things (IIoT) edge devices demands extreme compression, yet existing structured pruning methods collapse at high compression ratios due to one-shot importance estimation, and their cross-architecture behavior remains unpredictable. This article presents a cascaded multi-granularity pruning framework that removes layers, attention heads, and feed-forward channels in coarse-to-fine order, with lightweight low-rank recovery between stages to re-estimate component importance. An information-theoretic analysis motivates this ordering, and the Structural Independence Assumption (SIA) is formalized as a checkable condition predicting whether per-component pruning criteria are reliable for a given architecture: Multi-Head Attention (MHA)+GELU designs satisfy the SIA, whereas Grouped Query Attention (GQA)+SwiGLU designs violate it. On bearing fault diagnosis spanning 88M to 6.25B-parameter models, the framework extends achievable compression to 13.8 times on MHA+GELU architectures with 83.82% accuracy (+3.70 percentage points (pp) over the strongest baseline), while exposing a ~74pp accuracy collapse on GQA+SwiGLU architectures that violate the SIA. Deployed on an industrial slewing bearing fault diagnosis platform with NVIDIA DGX Spark, compressed models reduce inference latency by up to 67.2% and peak memory by 62.5%, demonstrating viability for IIoT edge inference.

Jinghan Wang, Yanjun Chen, Wei Zhang +3

School of Mechanical and Electrical Engineering, Harbin Institute of Technology, Harbin 150001, China · School of Information Science and Technology, Eastern Institute of Technology, Ningbo, China

LLM Inference Optimization Edge Devices Internet Of Thing Peak Memory

Jun 24, 2026cs.LG

Memory-Efficient Policy Libraries with Low-Rank Adaptation in Reinforcement Learning

When fine-tuning Large Language Models (LLMs), there has been success in minimizing both memory usage and computation with Parameter-Efficient Fine-Tuning (PEFT), like Low Rank Adaptation (LoRA). In this article, we have explored whether this approach is transferable to the world of robotics and Reinforcement Learning (RL), allowing learning with reduced memory usage and improved computational performance. Specifically, we focused on a version of multi-task robotics, where a library of specialist policies are created. In such a library memory efficiency is especially important. We used a Proximal Policy Optimization (PPO) algorithm and fine-tuned a baseline model to different tasks using LoRA. Our results demonstrate that, depending on the hyperparameters, LoRA can minimize memory usage by a factor of 20-160 compared to full fine-tuning of all layers. This implies a 90-95% storage saving when deploying a library of many (10-50) specialized policies, which can be the differentiating factor between being able to store the entire library in memory or having to use swap-memory in an applied robotics setting. At the same time, our results indicate that there is no significant difference in the success-rate between full fine-tuning and LoRA fine-tuning for the selected tasks.

Samuel Valland Lyngset, Tor Viljen Raanaas, Gard Sveipe +4

Department of Informatics, University of Oslo

Reinforcement Fine-Tuning Model Fine-Tuning Peak Memory

Jun 24, 2026cs.CV

Towards a Dynamic and Fixed-budget Memory Bank for Efficient Streaming Video Understanding

Currently, streaming video understanding is still a daunting task for existing \emph{multimodal large language models} (MLLMs). Its difficulties not only lie in handling the ever-increasing video frames, but also in the unpredictability of future video content and input instructions. In this paper, we study this task from the perspective of constructing a dynamic but fixed-budget memory bank, and propose a novel and training-free approach termed \emph{\textbf{CausalMem}}. CausalMem is dedicated to constructing a dynamic visual memory update mechanism, thereby maximizing the amount of information in streaming video within a limited memory space, much like the human brain. In practice, CausalMem estimates the redundancy of visual tokens and updates the memory bank via an online semantic basis, which models the principal semantics of the observed video stream. To validate CausalMem, we apply it to two representative MLLMs, namely LLaVA-OneVision and Qwen2.5-VL respectively, and conduct extensive experiments on both streaming and offline video understanding benchmarks. The experimental results not only show the great advantages than existing methods under both streaming and offline settings, \emph{e.g.},

+3.2\%

and

+3.0\%

average accuracy gains respectively, but also witness the superior semantic preservation for streaming videos, \emph{e.g.}, using 12

k

token budgets to memorize hour-long streaming videos, which achieves more than \textbf{20

\times

} visual token compression ratio and only occupies about \textbf{82 MB} storage. \textbf{Our code} is given in CausalMem.

Baiyang Song, Yuli Lin, Qiong Wu +5

1Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, 361005, P.R. China

Streaming Video Understanding Video Understanding Multimodal Large Language Models Peak Memory

Jun 23, 2026cs.CV

Pocket-SLAM: Rendering-Area-Aware Pruning for Memory-Efficient 3DGS-SLAM

3D Gaussian Splatting (3DGS) has garnered significant attention in Simultaneous Localization and Mapping (SLAM) due to its advances in capturing fine-grained geometry features and synthesizing novel views. For SLAM in large-scale scenes, such as autonomous driving, 3DGS-SLAM faces a critical limitation: memory consumption increases continuously over time as Gaussian points accumulate, leading to poor memory efficiency and limiting its applicability. In this work, we propose a rendering-area-aware pruning strategy that selectively removes Gaussians based on their contribution to the effective rendering area, rather than solely relying on Gaussian-level heuristics such as opacity or gradient magnitude. This perspective directly targets the sources of memory redundancy, effectively reducing the peak memory footprint of 3DGS-SLAM during runtime. Evaluations on the EuRoC and KITTI datasets demonstrate that our method consistently outperforms existing pruning approaches in large-scale outdoor scenes, achieving over 60% memory reduction and more than 2 times FPS improvement while preserving localization and mapping accuracy. These results highlight rendering-area-aware pruning as a promising direction for scaling 3DGS-SLAM to real-world autonomous driving scenarios. Our code is publicly available at https://github.com/UMN-ZhaoLab/Pocket-SLAM.git.

Leshu Li, Jie Peng, Yang Zhao

University of Minnesota, Twin Cities, USA · University of North Carolina at Chapel Hill, USA

Gaussian Splatting Simultaneous Localization And Mapping Systems Orb-Slam2 Autonomous Driving Peak Memory Renderer

Jun 23, 2026cs.CL

Metis: Bridging Text and Code Memory for Self-Evolving Agents

Self-evolving agents improve over time by distilling experience from past executions and reusing it in future tasks. Existing systems represent such experience either as natural-language text injected into the agent context or as code exposed as callable tools. However, the choice between these representations is typically made at design time rather than derived from the characteristics of the experience itself, leaving the trade-offs between them poorly understood. We present the first controlled study that isolates text memory and code memory over an identical set of experiences. Our results show that the two forms exhibit complementary trade-offs in construction cost, execution efficiency, and transferability, such that neither representation alone is sufficient. Guided by these findings, we propose Metis, a self-evolving agent system built on a hierarchical dual-representation memory. Metis organizes textual experience into execution plans, environment facts, and common pitfalls, and selectively crystallizes recurring plans into validated callable tools. This design combines the broad applicability of text memory with the execution efficiency of code memory while incurring tool-generation cost only when justified by repeated reuse. We evaluate Metis on AppWorld, a challenging benchmark for interactive agents. The results show that Metis improves task accuracy by up to 20.6% over ReAct while reducing execution cost by up to 22.8%. Compared with representative self-evolving agent systems, Metis consistently achieves a better balance between accuracy, execution efficiency, and memory-construction cost.

Zijie Dai, Siuhin He, Hui Li +9

The Chinese University of Hong Kong · Huawei · Wuhan University

Self-Evolving Agents Peak Memory Text Analysis and Detection

Jun 23, 2026cs.LG

RoPE-Aware Bit Allocation for KV-Cache Quantization

Existing low-bit KV-cache quantizers often treat each cached key as a flat vector. Under RoPE, however, a key's contribution to a future attention logit decomposes into a position-dependent sum over two-dimensional frequency blocks. This makes key-cache quantization a block-wise bit-allocation problem: high-energy RoPE blocks are more sensitive to quantization error and should receive more bits. We introduce Block-GTQ, a RoPE-aware bit allocator for key-cache quantization built on TurboQuant-MSE(TQ-MSE). For each layer and KV head, Block-GTQ computes a label-free energy score for each RoPE block and greedily allocates integer bit widths by marginal gain. Under matched K/V bit budgets, Block-GTQ better preserves RoPE query-key logits on a ten-model diagnostic panel, cutting per-layer MAE by 32-80% at 2 and 3 b/dim K-only quantization and winning all 367/367 layer comparisons against uniform TQ-MSE. These fidelity gains translate to stronger downstream long-context retrieval, understanding, and reasoning. At K2V2 on Llama-3.1-8B-Instruct, Block-GTQ raises the six-task NIAH average from 70.6 to 97.4, and the LongBench-EN average from 36.87 to 53.31. On AIME 2024/2025 with DeepSeek-R1-Distill-Qwen-7B, without an fp16 recent-key buffer, Block-GTQ at K3V2 scores 51.7/37.5, close to fp16's 54.2/37.9, whereas uniform TQ-MSE collapses to 0.0/0.0. We further implement a packed-cache serving path. On a single H800 GPU with Qwen2.5-3B-Instruct, packed K3V3 achieves 3.24x KV-cache compression with fp16-comparable quality, runs 1.34x faster than fp16 FlashAttention2 at 128K context, reduces peak memory from 56.31 GB to 19.85 GB, and remains feasible at 256K and 512K where fp16 OOMs. Code is available at https://github.com/JIA-Lab-research/blockgtq.

Fengfeng Liang, Yuechen Zhang, Jiaya Jia

1Hong Kong University of Science and Technology · 2The Chinese University of Hong Kong · 3MiMo, Xiaomi Corporation

Key-Value Cache Quantization Peak Memory Deepseek Mean Squared Error

Jun 22, 2026cs.LG

FlowTrain: Flow-Based Decoupled Training for Industrial-Grade Vision-Language Models

Industrial-grade distributed training of vision-language models (VLMs) remains far less efficient than that of unimodal LLMs. Existing solutions either follow a monolithic design that assigns uniform parallelism to heterogeneous modules or adopt a disaggregated deployment that separates modules while executing them as a batch-synchronized pipeline. In this paper, we highlight that the above solutions are still not sufficient, and VLM training can be further decoupled. To this end, we present FlowTrain, a flow-based decoupled training framework that reformulates VLM training as a producer-consumer dataflow coordinated through a unified memory pool. The encoder and backbone can progress independently over a global virtual address space. Since this execution decoupling fundamentally changes the optimization objective of allocation and scheduling, FlowTrain further introduces a heterogeneous parallel allocator that assigns module-specific parallelism strategies by solving a throughput matching problem. The dynamic packing scheduler is used to construct balanced microbatches at runtime according to the actual LLM-side computation cost. Extensive experiments on real-world workloads show that FlowTrain achieves over 50% MFU and up to 1.7x throughput improvement, narrowing the efficiency gap to LLM-only training.

Zhida Jiang, Zhaolong Xing, Yang Pei +14

Huawei

Recent Vision-Language Models Peak Memory

Jun 22, 2026cs.RO

Memory-Native Non-Terrestrial Networks for Embodied Intelligence

Non-terrestrial networks (NTN) provide ubiquitous connectivity for embodied intelligence (EI), enabling robots in wilderness to leverage cloud resources or report critical information to remote centers. However, the synergy is nontrivial due to the highly-dynamic, resource-constrained, topology-varying, and task-oriented environment. Existing memoryless NTN protocols become inefficient, since the decisions are driven by local channel conditions and instantaneous service demands. To address these limitations, this paper proposes the memory-native NTN (MemNTN) paradigm that leverages long-horizon contexts for memory augmented system optimization. To realize this paradigm shift, we establish a dual-memory architecture that distinguishes between physical memory representing the state of the world and digital memory encoding historical network experience. We develop memory acquisition, compression, valuation, update, and utilization mechanisms that facilitate cross-layer, memory-native decision-making, spanning from the physical and access layers up to the network and application layers. Experiments in satellite embodied question answering (SEQA) demonstrate that the proposed MemNTN significantly outperforms conventional stateless NTN and terrestrial approaches.

Chengyang Li, Yikun Wang, Jiahui He +6

The University of Hong Kong, Hong Kong, China · Southern University of Science and Technology, Shenzhen, China · Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China +2

Embodied Artificial Intelligence Peak Memory

Jun 19, 2026cs.PF

Does Mixture-of-Experts Actually Help Inference on Consumer and Edge Hardware? An Empirical Study

Mixture-of-Experts (MoE) language models are often described as ideal for resource-constrained inference. Each token activates only a small subset of experts, so the per-token compute cost, in floating-point operations (FLOPs), resembles that of a much smaller dense model. Whether that FLOP advantage survives in practice is far less clear. We ask whether MoE models actually run faster and cheaper than comparable dense models on consumer-grade and edge hardware. We benchmark OLMoE-1B-7B (1.3 B active of 6.9 B total) against three dense baselines on an Apple M2 Pro and an NVIDIA Jetson Orin Nano 8 GB through llama

.

cpp, measuring throughput, memory, and on-device energy. The answer is device-dependent: OLMoE's active-parameter advantage is only partly realised on the laptop (~10% behind the same-active Llama-3.2-1B) and erodes on the edge device (~31% behind, at 2.1

\times

the energy per token, with peak memory at the 8 GB ceiling). Patching llama

.

cpp to time the decode graph node-by-node shows routing accounts for under 9% of MoE-block compute on the cleaner edge backend, so the gap reflects total-parameter memory footprint, expert dispatch, and KV-cache pressure rather than routing. The implication is that on bandwidth-bound edge hardware, inference cost tracks total parameters, not active ones, and sparse activation does not buy back what the device is constrained on. These findings are bounded to one MoE model at this parameter scale and two devices, and we release the full measurement harness and per-run data.

Alfarizy Alfarizy, Hung Truong Thanh Nguyen, René Richard +2

Analytics Everywhere Lab, University of New Brunswick, Fredericton, NB, Canada · TSAI Lab, University of New Brunswick, Fredericton, NB, Canada

Mixture-Of-Expert Inference Mixture-Of-Experts Localized Lora-Moe Inference Cost Edge Hardware Peak Memory+2

Jun 18, 2026cs.RO

MemoryWAM: Efficient World Action Modeling with Persistent Memory

Robust robotic manipulation in the real world requires not only an understanding of the current observation, but also memory and dynamics modeling. World action models (WAMs) possess these capabilities by jointly modeling visual foresight and actions conditioned on both current and historical observations, making them a promising paradigm for robotic manipulation. However, existing WAMs face a fundamental trade-off: methods with efficient inference typically condition only on a bounded window of recent observations and therefore struggle in non-Markovian environments, whereas methods that preserve long histories incur time and space costs that grow substantially with sequence length. To address this challenge, we introduce MemoryWAM, a world action model with efficient persistent memory. MemoryWAM uses a hybrid memory design that combines recent frames, event-boundary anchor frames, and compact gist tokens that summarize long-range history. A tailored attention mechanism enables retrieval of both detailed short-term context and compressed long-term context, supporting memory-dependent decision-making with reduced inference latency and GPU memory usage. Across long-horizon, memory-dependent manipulation tasks in both simulation and the real world, MemoryWAM outperforms strong vision-language-action (VLA) and WAM baselines while maintaining favorable computational efficiency.

Sizhe Yang, Juncheng Mu, Tianming Wei +8

The Chinese University of Hong Kong · Tsinghua University · Zhejiang University

Efficient World-Action Model World Models Peak Memory World Foresight Efficient Inference

Jun 18, 2026cs.SE

A Topology-Aware, Memory-Centric Architecture that Separates Root-Cause Derivation from Root-Cause Explanation

Modern microservice deployments fail in ways that are easy to detect and hard to explain. When a fault propagates along service dependencies, alerts fire in floods, dashboards multiply, and the scarce resource, an engineer who understands how the services relate, is consumed reconstructing context that the monitoring stack discarded. We argue that the missing ingredient in autonomous operations is not a better anomaly detector or a larger language model, but operational memory: a persistent, structured representation of how a system normally behaves, how its parts depend on one another, and how it has failed before. We present O PS C ORTEX, a working multi-agent prototype that organizes this memory into four tiers and uses it to separate two tasks the field usually conflates: deriving a root cause and explaining it. Root cause is computed deterministically from a learned dependency graph and the temporal ordering of threshold crossings; a large language model (LLM) is then asked only to explain, confirm, and recommend, using evidence the system has already assembled. We motivate the design with two documented production cascading failures, review representative literature on observability, anomaly detection, graph-based localization, and LLM-assisted diagnosis, and show how each architectural choice maps directly to a failure mode those incidents exhibit. The prototype is validated on an instrumented e-commerce benchmark with eight injectable failure scenarios.

Momil Seedat

Independent Research Prototype

Microservices Root Cause Analysis Multi-Agent Communication Peak Memory

Jun 18, 2026cs.CL

AtomMem: Building Simple and Effective Memory System for LLM Agents via Atomic Facts

Large language models (LLMs) demonstrate strong reasoning and generation abilities, but their fixed context windows limit long-term information accumulation and reuse across multi-session interactions. Existing memory-augmented systems often construct memory in a coarse and unstable manner, relying on inefficient memory representations or unstable unconstrained updates. To address these challenges, we propose AtomMem, a long-term memory system designed for value-dense storage and stable memory evolution. AtomMem introduces a Fact Executor, which selectively extracts high value atomic facts from long form interactions to serve as highly efficient memory representations. Subsequently, AtomMem organizes these facts into hierarchical event structures and temporal profiles, capturing coherent episodic contexts and tracking dynamically evolving user attributes over time. During retrieval, the system activates an associative memory graph to connect fragmented memories. Experiments on the LoCoMo benchmark confirm that AtomMem achieves state-of-the-art performance across various reasoning tasks, offering a scalable and economically viable solution for deploying intelligent personalized agents.

Yanyu Yao, Shangze Li, Zhi Zheng +4

State Key Laboratory of Cognitive Intelligence, University of Science and Technology of China, Hefei, China · Anhui University, Hefei, China

Peak Memory Large Language Model Agents Multi-Turn Interactions Long-Term Atoms

Jun 17, 2026cs.LG

Techniques for Peak Memory Reduction for LoRA Fine-tuning of LLMs on Edge Devices

Fine-tuning of Large Language Models (LLMs) using Low-Rank Adaptation (LoRA) on an end-user's data offers personalized experiences while keeping data private, but faces severe memory constraints on consumer hardware. Peak memory during fine-tuning often exceeds device limits, especially for models with billions of parameters and long-context training data. This paper introduces a suite of complementary techniques to reduce memory footprint without sacrificing model quality: (1) base model quantization with on-the-fly dequantization, (2) memory-efficient checkpointing combining selective activation caching and disk offloading, (3) softmax approximation using semantically relevant token subsets, and (4) logits masking. Experiments on Llama-3.2 3B and Qwen-2.5 3B demonstrate up to

26\times

and

28\times

reduction in peak memory, enabling fine-tuning on resource-constrained devices.

Hassan Dbouk, Matthias Reisser, Prathamesh Mandke +2

Qualcomm AI Research is an initiative of Qualcomm Technologies, Inc. · work done while at Qualcomm

Large Language Model Fine-Tuning Model Fine-Tuning Peak Memory Edge Devices Quantizer

Jun 17, 2026cs.AI

Repeated Shared Access Enables Grokking, but Edit Propagation Depends on an Addressable Memory

We study factual edit propagation in a controlled synthetic knowledge-graph QA setting using a 2x2 grid that crosses loop recurrence with shared-memory access: a dense transformer (Dense), a looped transformer (Loop), a dense backbone with shared memory (Dense+Mem), and a looped backbone with shared memory (loop-memory coupling, LMC). The two factors dissociate. For learning, both routes to repeated shared access -- looped recomputation and repeated memory rereading -- cross the out-of-distribution (OOD) grokking barrier that Dense fails, so repeated shared access is the behavioral regularity, not a specific architecture. For editing, the substrates split along a different axis: applying a single localized factual edit (conditioned on direct success) and measuring 2-hop propagation on a shared pre-edit-correct set, the edit propagates strongly in both memory-bearing cells (LMC 0.78-0.92, Dense+Mem 0.71-0.96) and only weakly in the memory-free ones (Loop 0.04-0.30, Dense 0.00-0.03). The split is along the memory axis, not the loop axis: every memory-bearing seed exceeds every memory-free seed, with no detectable difference between the two memory cells. Crucially Dense+Mem has no recurrence, so the propagating ingredient is an addressable site that an edit can write to and later computation rereads, not loop recomputation; Loop is at best a partial intermediate. The affordance survives coarsening the store (N=128 to N=13): propagation attenuates but the memory/no-memory split persists, so fine granularity buys precision rather than the affordance itself. These results dissociate learning competence from editing affordance -- repeated shared access suffices to grok, but edit propagation depends on whether the substrate exposes an addressable memory that the forward computation can write to and later reread, an affordance that loop recurrence provides only partially.

Yanan Niu

Kuaishou Technology

Peak Memory Shared Control Knowledge Graphs Transformer Architectures Affordances

Jun 17, 2026cs.CV

How Well Can Your Video Model Remember? Measuring Memory-Budget Trade-offs in Long Video Understanding

We introduce a compact empirical model that quantifies how answer accuracy degrades as a function of frame budget B and temporal distance D in long video understanding -- analyzing performance when recalling content from D seconds in the past using a fraction B of total frames. Long-form models operate under strict budgets, yet no prior framework predicts how accuracy degrades as B shrinks and events recede. We fit a weighted least-squares model on ~155,000 binary predictions across ten models and three sampling strategies, deriving a law where logit-accuracy scales linearly in log-budget with a distance-dependent exponent that decays log-linearly with distance. This budget exponent α(D) captures the marginal value of extra frames at distance D. The law achieves cell-level weighted R^2 = 0.05-0.75 across models. Notably, budget effectiveness at D = 1000 s differs by \approx 7.4\times between the best streaming and base models. STREAMINGVLM achieves α(1000) = 1.26 (95% CI: [1.06, 1.58]), meaning a tenfold budget increase substantially improves long-distance accuracy, while the best Qwen3-VL base model reaches only α(1000) = 0.17 (CI: [0.04, 0.34]). In accuracy space, a 10\times budget increase at D = 1000 s yields +29 percentage points for STREAMINGVLM versus +4 pp for the base model. Sampling strategies show model-dependent trade-offs: random sampling yields higher base sensitivity but steeper distance decay. We demonstrate how α(D) enables principled budget allocation, including a model-ranking reversal at long distance, and propose it as a diagnostic metric for streaming video models.

Yixian Tian

1Fudan University · 2Beijing University of Posts and Telecommunications

Video Understanding Token Budget Allocation Peak Memory

Jun 16, 2026cs.CL

CoreMem: Riemannian Retrieval and Fisher-Guided Distillation for Long-Term Memory in Dialogue Agents

Personalized dialogue agents require continuous long-term memory to maintain coherent interactions across multiple sessions. However, deploying these capabilities on consumer-grade hardware (e.g., 8 GB VRAM edge devices) introduces severe memory and compute bottlenecks. Existing systems typically rely on isotropic cosine similarity for retrieval and heuristic rules for context compression. These approaches lack a unified theoretical foundation, frequently suffering from the hubness problem in high-dimensional retrieval and syntactic fragmentation during compression. To overcome these limitations, we propose CoreMem, a resource-efficient edge-cloud memory architecture fundamentally unified by information geometry. First, Riemannian retrieval replaces cosine matching with a locally adaptive Fisher-Rao metric, effectively penalizing hub memories via Mahalanobis distance with O(Ndr) Woodbury acceleration for real-time search. Second, Fisher-guided discrete token distillation (FDTD) introduces a hierarchical sentence-to-token compression mechanism. It derives sensitivity scores from Fisher information traces, providing a principled compression-KL tradeoff augmented with explicit structural syntax protection. Evaluated on the LOCOMO and LongMemEval-S benchmarks, CoreMem achieves strong accuracy improvements, yielding substantial gains in Open-domain (+4.51 pp) and Temporal (+4.17 pp) reasoning. Extensive profiling confirms that CoreMem operates seamlessly within a strict 8 GB VRAM budget, successfully bridging the gap between resource-constrained edge devices and the demand for theoretically grounded, lifelong memory agents.

Jiaqi Chen, Yongqin Zeng, Shaoshen Chen +4

Shenzhen International Graduate School, Tsinghua University · Peng Cheng Laboratory · Shandong Analysis and Test Center, Qilu University of Technology, Jinan, China +1

Peak Memory Long-Term Memory Token Compression

Jun 16, 2026cs.LG

Ternary Mamba: Grouped Quantization-Aware Training of W1.58A16 State Space Models

State Space Models (SSMs) such as Mamba-2 offer linear-time inference but their memory footprint limits edge deployment. Prior ternary SSM work (Slender-Mamba) trains from scratch on 150B tokens; we show a pretrained checkpoint suffices, reducing the marginal token budget by 1,000x. Using grouped quantization-aware training (QAT) with knowledge distillation from a frozen FP16 teacher, we compress Mamba-2 1.3B to 3.61x (2,687 to 744 MB) and achieve 48.1% zero-shot accuracy (7-task average) in just 102M tokens (4 GPU-hours, single H100) -- approaching Bi-Mamba's 48.4% (within +/-0.9pp CI). This QAT-from-pretrained setting reveals zero-ratio collapse, a novel instability caused by learnable quantization scales that does not arise in from-scratch training. We further show that post-hoc correction strategies effective for Transformers fail for SSMs due to error accumulation through the recurrence. These results demonstrate that ternary SSMs do not require expensive from-scratch training: QAT from pretrained checkpoints with KD is a data-efficient alternative.

Ramprasath Ganesaraja, Sahil Dilip Panse, Swathika N

EdgeVerve Systems Limited

Quantization-Aware Training State Space Models Peak Memory

Jun 15, 2026cs.AI

User as Code: Executable Memory for Personalized Agents

A personalized AI agent needs a user memory: a persistent model of who the user is, built across many conversations and consulted on each new one. Today this memory is almost always stored as unstructured text, a knowledge graph, or a flat store of facts, and consulted by retrieval -- fetching the entries most similar to the current request. Such "bag-of-facts" memory recalls individual facts well, but because storing a fact and acting on it are separate steps, it struggles to resolve contradictions, aggregate over many records, or enforce rules. We argue that user memory should instead be executable. We introduce User as Code (UaC), a paradigm in which an agent's model of a user is a living software project: typed Python objects hold the user's state and ordinary Python functions encode the rules that govern it, so representing and reasoning about the user happen in one medium an interpreter can run. The enabling mechanism is a two-phase pipeline: an append-only log that never discards a fact, periodically checkpointed into typed code. This changes what memory can do. On standard long-term conversation benchmarks, UaC matches both a full-context upper bound and the strongest prior memory systems on recall (78.8% on LOCOMO). Its advantage emerges where representation matters most. On aggregate questions over a user's history -- "how many international trips did I take last year?" -- retrieval-based memory collapses (6-43%) while UaC stays near-perfect (99%), because the answer is a one-line computation over typed state rather than a search over text. And because its rules execute deterministically whenever the state changes, UaC can surface unsolicited, safety-critical alerts -- such as a newly prescribed drug that conflicts with an allergy recorded months earlier -- a capability query-driven memory cannot provide.

Bojie Li

Pine AI

Artificial Intelligence Agents Peak Memory

Jun 15, 2026cs.CL

HiMPO: Hindsight-Informed Memory Policy Optimization for Less-Entangled Credit in Long-Horizon Agents

Long-horizon agents rely on memory mechanisms to compress interaction history, but optimizing memory writing faces a distinct credit assignment challenge: a memory update may be rewarded or penalized due to downstream tool failures, noisy observations, or reasoning errors rather than its own contribution. We propose HiMPO, a Hindsight-Informed Memory Policy Optimization framework for assigning less-entangled credit to memory-writing actions in long-horizon agents. HiMPO first estimates the local utility of a memory update by comparing the task-relevant information recoverable from the previous and updated memories under the same pre-write state. It then uses hindsight relevance as a bounded retrospective filter that attenuates memory credit when local utility is not supported by the target outcome. The resulting memory-specific advantage is applied only to memory tokens, while trajectory-level rewards optimize the rest of the agent's behavior. Across judge-based open-domain tasks and objective compressive-memory QA, HiMPO improves over strong memory-based and RL-based baselines while preserving compressed-context efficiency. Controlled interventions and live replay studies further show that HiMPO reduces blame leakage from tool-induced errors, assigns memory credit that aligns with the functional impact of memory writes, and remains robust to noisy training targets.

Jiangze Yan, Yi Shen, Wenjing Zhang +5

Unicom Data Intelligence, China Unicom · Data Science & Artificial Intelligence Research Institute, China Unicom

Long-Horizon Agents Long-Term Agent Memory Hindsight Credit Assignment Step-Level Credit Peak Memory

Jun 12, 2026cs.CL

Are Online Skill and Memory Modules Always Worth Their Tokens? A Budget-Constrained Study of Web Agents

Online web agents often augment a base actor with memory, workflow, or skill modules. These modules can improve performance, but they also consume test-time tokens, a cost rarely reported alongside the actor's inference cost. We study online augmentation, where this overhead is paid on every task, and re-evaluate its benefits under a fixed total inference budget. We compare AWM, ASI, and ReasoningBank with a token-matched vanilla baseline that uses the same budget for additional actor steps. Across four WebArena domains and three models, Gemini 3 Flash, GPT-5.4-mini, and Qwen 3.6-27B, the vanilla baseline matches or surpasses all three augmentation methods in aggregate success rate while often using fewer total tokens. We observe a similar trend on WorkArena-L1 with Qwen 3.6-27B, indicating that the effect extends to enterprise knowledge-work tasks. Our results suggest that skills and workflow memory can be useful in specific domains, but their apparent gains often vanish against a budget-matched actor. We further show that run-to-run variance materially affects outcomes and should be reported as a core evaluation criterion for online web agents.

Sina Hajimiri, Masih Aminbeidokhti, Jose Dolz +4

ServiceNow AI Research · ÉTS Montreal · University of British Columbia +1

Token Budget Allocation Skills Inference Cost Peak Memory Webarena Domains Agentic Workflow Design

Jun 12, 2026cs.LG

When to Write and When to Suppress: Route-Specialized Dual Adapters for Memory-Assisted Knowledge Editing

Knowledge editing systems must update selected facts while preserving nearby but irrelevant behavior. This paper studies this problem in a memory-assisted setting where an edit memory is retrieved at inference time and a parameter-efficient adapter corrects the model's object preference. We argue that the central design question is not only how to write an edit, but also when to suppress it. We introduce RRDA, a route-specialized dual-adapter editor. A relevance router first decides whether a prompt should receive an edit memory. Routed prompts use an edit adapter trained to prefer the new object over the original object; unrouted non-direct prompts use a separate locality adapter trained to preserve or restore the original-object preference. We evaluate RRDA on three 1,000-case protocols, CounterFact, ZsRE, and MQuAKE-CF, under the same memory protocol and two 7B/8B base models. On Llama-3.1-8B-Instruct, RRDA obtains the best overall probability-preference accuracy on all three benchmarks: 0.8180 on CounterFact, 0.8946 on ZsRE, and 0.9922 on MQuAKE-CF. The same trend holds on Qwen3-8B. Router ablations show that the relevant memory boundary differs across datasets: a lexical neural router is safest on CounterFact, while BGE embedding routing is better on ZsRE and MQuAKE-CF. Memory, component, and module ablations show that explicit memory supplies the largest edit gain, while route-specialized adapters improve the final reliability-locality balance rather than simply increasing LoRA capacity.

Baijia Zhang, Yining Huang

Xinjiang University · Meta Emergence Laboratory

Multimodal Knowledge Editing Peak Memory Suppression Preference Alignment Learning

Jun 12, 2026cs.LG

Semidirect Fourier Delta Attention: Phase-Controlled Delta Memory with Constructive Chunk-WY Kernels

Linear attention replaces softmax attention's growing KV cache with a fixed recurrent state, but this compression limits exact state tracking and long-context memory. We introduce \emph{Semidirect Fourier Delta Attention} (SFDA), a phase-controlled generalization of Kimi Delta Attention that replaces real diagonal decay with block-rotational Fourier control:

S_t=(I-β_t k_tk_t^*)Λ_tS_{t-1}+β_tk_tv_t^*, \qquad Λ_t=\diag(α_t\odot e^{iθ_t}).

Our main result is a constructive chunk-WY factorization for products

A_t=Λ_t-u_tr_t^*

, giving

A_t\cdots A_1=Γ_t-Y_tM_tW_t^*

with rank growth bounded inside fixed chunks. This yields an exact affine chunk transfer, formal stability and complexity bounds, and a compact characterization of phase-plus-low-rank memory. We verify the algebra numerically and show in toy state-tracking experiments that SFDA learns cyclic memory where the phase-disabled KDA baseline remains near chance. Fused kernels and large-scale language-model comparisons are left to future work.

Tiantian Zhang

Columbia University

Kimi Delta Attention Recurrent State Peak Memory Low-Rank Structure

Jun 11, 2026math.OC

Scalable Deep Unfolding of Conic Optimizers

Deep unfolding (DU) accelerates iterative optimizers by introducing learnable components and training them through unrolled iterations, but extending DU to the large-scale semidefinite programs (SDPs) common in robotics has remained limited. Unrolling a full-update conic solver such as COSMO exposes two obstacles that prior work on learned conic solvers has not: backpropagating through the per-iteration linear-system solve incurs memory quadratic in the problem size once the coefficient matrix is formed explicitly, and backpropagating through the positive semidefinite (PSD) cone projection becomes numerically unstable when eigenvalues coincide. We address the first obstacle with a matrix-free implicit differentiation rule that operates entirely through matrix-vector products, reducing memory from

O(n^2)

to

O(n)

and enabling backpropagation at scales where direct factorization runs out of memory. We address the second with a backward rule based on the Dalečkii--Krein representation of the Fréchet derivative, which remains well-defined under repeated eigenvalues. Together these make it possible to learn lightweight hyperparameter policies and warm-starts for a full-update conic solver. We evaluate on nonlinear covariance steering problems solved via sequential convex programming (SCP), as well as standalone SDPs and second-order cone programs ranging from max-cut and Lovász

\vartheta

SDPs to robust estimation and control problems. The learned policies outperform state-of-the-art solvers across all problems, and can provide up to a 50

\times

speedup depending on the class. When used as a subroutine in SCP, the learned approach delivers over a 30

\times

speedup compared to COSMO.

Alex Oshin, Rahul Vodeb Ghosh, Evangelos A. Theodorou

Georgia Institute of Technology

Deep Unfolding Networks Convex Relaxation Covariance Peak Memory

Jun 11, 2026cs.LG

How Much Memory Do We Need? Adaptive Memory Gate for Neural Operators

Neural operators have emerged as a powerful data-driven approach for solving time-dependent PDEs. Among recent advances, memory-augmented neural operators explicitly incorporate past states and have achieved remarkable performance under low-resolution observation settings. However, existing approaches apply a fixed memory weight regardless of observation conditions, such as resolution or physical parameters, limiting their adaptability. Our preliminary experiments reveal that optimal memory weight varies with resolution and viscosity, implying that a fixed memory weight cannot simultaneously optimize performance across diverse settings. We propose AMGFNO, which dynamically modulates memory weight through a learnable gate. On the Kuramoto-Sivashinsky and Burgers' equations, AMGFNO achieves 55-79% nRMSE reduction over at low resolution, with the learned gate value automatically decreasing from

\bar{g} \approx 0.7

to near-zero as resolution increases.

Jihyeon Hur, Yongseok Kwon, Min-Gi Jo +2

1KAIST, Republic of Korea.

Neural Operators Neural Partial Differential Equation Solvers Neural Network Optimization Partial Differential Equations Peak Memory Hilbert Series

Jun 11, 2026cs.LG

WHAR Arena: Benchmarking the State of the Art in Efficient Wearable Human Activity Recognition

Deep learning has become the dominant paradigm in Wearable Human Activity Recognition (WHAR), yet progress is obscured by a comparability crisis. Results are often reported using inconsistent datasets, custom data processing, and varying evaluation protocols, making state-of-the-art claims fragile. We address this with a large-scale, open-source benchmark that integrates 30 diverse datasets under standardized processing, unified model interfaces, and a shared cross-subject evaluation protocol. Evaluating 17 representative architectures across 4760 training runs, we jointly measure predictive performance alongside on-device latency, peak memory, and model size on an Android reference device. Our results reveal that the WHAR state of the art is distributed rather than dominated by a single architecture. While CNN-HAR achieves the highest mean macro-F1, top-performing models cluster tightly, indicating contemporary architectures have converged near a predictive performance ceiling. When accounting for deployment efficiency, compact neural models, such as TinierHAR, and classical Random Forests define the practically relevant Pareto frontier, whereas larger recurrent and hybrid models incur high hardware costs without corresponding performance gains. Consequently, while predictive performance has plateaued, substantial potential for future progress remains in optimizing deployment efficiency and improving adaptation to domain shifts. We release our full framework to support transparent reuse and extension.

Maximilian Burzer, Tobias King, Till Riedel +2

Karlsruhe Institute of Technology, Karlsruhe, Germany · IPAI Foundation gGmbH, Heilbronn, Germany

Human Activity Recognition Deep Learning Evaluation Protocols Peak Memory

Jun 11, 2026cs.CL

MemRefine: LLM-Guided Compression for Long-Term Agent Memory

Large language model (LLM) agents are increasingly expected to operate over long-term interactions, where information from past dialogues must be preserved and recalled to support future tasks. However, as interactions accumulate, the memory store grows without bound and fills with redundant entries that inflate storage cost and degrade retrieval by crowding out the most useful evidence. Furthermore, this is especially limiting on resource-constrained platforms with hard memory budgets, motivating us to formulate storage-budgeted memory management, the task of keeping an already constructed memory store within a fixed budget while preserving information useful for future interactions. To this end, we then propose MemRefine, an LLM-guided framework that, since surface similarity poorly reflects factual value, uses similarity only to propose candidate pairs and defers delete, merge, and preserve decisions to an LLM judge based on factual content, iterating until the budget is met. Across multiple memory frameworks and long-term conversation benchmarks, MemRefine consistently meets target budgets while preserving downstream performance and outperforming rule-based baselines under tight budgets.

Minjae Kim, Jinheon Baek, Soyeong Jeong +1

Korea University · KAIST

Long-Term Agent Memory Large Language Model Memory Peak Memory Redundancy Data Compression Methods Refinevla

Jun 11, 2026cs.CL

G-Long: Graph-Enhanced Memory Management for Efficient Long-Term Dialogue Agents

While Large Language Models (LLMs) have advanced open-domain dialogue systems, maintaining long-term consistency remains a challenge due to inherent limitations in long-context reasoning and the inefficiency of processing extensive raw text. Existing approaches typically rely on either unstructured memory storage, which is prone to information loss, or computationally expensive LLMs that incur high latency. To address these limitations, we propose G-Long, a graph-enhanced framework that utilizes a fine-tuned small Language Model (sLM) for structured triplet extraction and associative retrieval, significantly reducing operational costs. Furthermore, we introduce the novel attention-aware importance scoring mechanism that leverages the intrinsic cross-attention signals of a T5 summarizer to identify salient memories. Extensive experiments across diverse benchmarks demonstrate that G-Long achieves state-of-the-art performance in both response generation and memory retrieval, yielding performance gains of up to 9.8% in response quality on MSC and 40.8% in retrieval recall on LME, while significantly minimizing computational overhead.

Minjun Choi, Yoonjin Jang, Sangwon Youn +1

Sungkyunkwan University, Suwon, Republic of Korea

Large Language Model Memory Peak Memory Long-Term Rdf

Jun 10, 2026cs.LG

On Subquadratic Architectures: From Applications to Principles

Transformers dominate modern sequence modeling, but their quadratic attention incurs substantial computational cost. Subquadratic architectures offer a scalable alternative. However, it remains unclear which designs yield the most effective sequence models. We compare three leading approaches: xLSTM, Mamba-2, and Gated DeltaNet. We evaluate these models on tasks with complex dependencies: (1) code-model pre-training, (2) distillation of code models from large language models, and (3) pre-training of time-series foundation models. Across these settings, xLSTM delivers the strongest overall performance. To explain xLSTM's advantage, we present a unified formulation and analyze the underlying architectural mechanisms, focusing on state tracking and memory dynamics. Our results show that xLSTM enables more flexible and stable memory correction via its gating scheme. We corroborate these findings on controlled synthetic length-generalization tasks. Overall, our findings indicate that xLSTM's gains on complex tasks stem from robust state tracking and accumulation.

Anamaria-Roberta Hartl, Levente Zólyomi, David Stap +6

Johannes Kepler University Linz · NXAI GmbH

Sequence Modeling Linear Attention Gated Deltanet Peak Memory

Jun 10, 2026cs.RO

DrivingAgent: Design and Scheduling Agents for Autonomous Driving Systems

Many autonomous driving systems are increasingly incorporating foundation models to improve generalization and handle long-tail scenarios. However, this trend introduces two key challenges: (i) the manual and labor-intensive process of designing and integrating new models, and (ii) the lack of intelligent, dynamic scheduling mechanisms to meet strict real-time constraints. While Large Language Model (LLM)-based agents offer a promising avenue for automation, existing frameworks are ill-suited for autonomous driving. Specifically, they fail to distinguish between the fundamentally different requirements of system design and real-time scheduling, treat modules as opaque black boxes, and are not designed for continuous operation. To address these limitations, we propose DrivingAgent, a novel agent framework tailored to the dual challenges of autonomous driving system design and scheduling. In the design phase, DrivingAgent automates module development by interpreting system architecture, generating code, and validating modules via super-network training. In the scheduling phase, it employs a lightweight LLM trained with reinforcement learning to dynamically orchestrate system modules in real time, supported by a structured memory that integrates long-term storage with timestamped short-term context. Experimental results demonstrate that DrivingAgent achieves a superior speed--accuracy trade-off on both the nuScenes and Bench2Drive benchmarks.

Zhongyu Xia, Wenhao Chen, Yongtao Wang +1

Wangxuan Institute of Computer Technology, Peking University · University of California, Merced

Autonomous Driving Schedule Agentic Framework Peak Memory

Jun 10, 2026cs.CL

Augmenting Molecular Language Models with Local $n$ -gram Memory

Transformer-based language models for SMILES strings suffer from a locality gap: standard character-level tokenization fragments chemically meaningful motifs, forcing models to repeatedly learn local syntax at the expense of long-range dependencies. To address this without disrupting standard tokenizers, we propose MolGram, which integrates a conditional

n

-gram memory module into molecular language models. MolGram maps local string patterns to learned embeddings via scalable hash lookups and dynamically injects this regional context into hidden states. Evaluations across three tasks, including unconditional molecule generation, forward reaction prediction, and single-step retrosynthesis, show that MolGram consistently improves performance. Crucially, our analyses demonstrate that MolGram outperforms baselines with 3

\times

more parameters, establishing explicit local pattern memory as a highly efficient inductive bias.

Xinni Zhang, Zijing Liu, He Cao +2

1The Chinese University of Hong Kong · 2International Digital Economy Academy

Unigram Tokenizer Language Modeling Tokenizer Peak Memory

Jun 9, 2026eess.AS

Massive Open-Vocabulary Keyword Spotting

Automatic speech recognition systems have been shown to under-perform when it comes to transcribing words rarely seen in the training data, namely specialized terminology. Open-vocabulary keyword spotting, combined with contextual biasing, has been shown to mitigate this issue. However, existing systems can only handle glossaries of a few hundred terms without becoming an infeasible bottleneck. We propose a system that stores features with a memory footprint up to 128 times smaller than a comparable baseline and allows users to process massive databases while remaining open-vocabulary. Without fine-tuning the speech recognition model, our system achieves a comparable entity recall as uncompressed solutions, even in languages not seen during training.

Leonor Barreiros, Raul Monteiro, Afonso Mendes +1

Priberam Labs, Lisboa, Portugal · Instituto Superior Técnico, Lisboa, Portugal · Instituto de Telecomunicações, Lisboa, Portugal

Keyword Spotting Automatic Speech Recognition Open-Vocabulary Training Data Peak Memory

Jun 9, 2026cs.LG

Unifying Data, Memory, and Compute Efficiency in LLM training: A Survey

Resource constraints increasingly determine what can be trained, fine-tuned, and deployed in large language models (LLMs), yet efficiency is often studied through isolated techniques rather than as an interacting system of limits. This survey adopts a constraint-centric perspective and organizes recent progress around three coupled bottlenecks: data efficiency (what to train on), memory efficiency (how to fit training), and compute budget awareness (when and where to spend FLOPs). On the data axis, we review selection and pruning methods that maximize learning per token, ranging from scalable proxy signals based on learning dynamics to gradient- and influence-based scoring, as well as difficulty-aware and curriculum-style strategies. We highlight emerging evidence that different notions of good data dominate in different regimes, implying that optimal subsets depend on the task objective and resource budget rather than being universal. On the systems side, we show that GPU memory, not raw compute, is often the dominant bottleneck in fine-tuning, and that effective scaling requires jointly reducing weight storage, optimizer states, and activation memory rather than optimizing any single component in isolation. Beyond memory, we frame training and inference as compute-governed processes in which optimization, data selection, and decoding must explicitly account for finite FLOP budgets. We review evidence for compute-optimal allocation and stopping rules, where computation should be halted or reallocated once marginal performance gains fall below a budget-dependent threshold. Together, these results unify compute-aware data selection, scaling laws, and adaptive inference under a common principle of resource-conditioned decision-making.

Vanessa Schmidt, Huy Hoang Nguyen, Cédric Jung +2

Chair of Information Theory and Data Analytics (INDA), RWTH Aachen University, 52074 Aachen, Germany · AIT Austrian Institute of Technology GmbH, 1210 Vienna, Austria · Automation and Control Institute, Technische Universität Wien (TUW), 1040 Vienna, Austria

Alignment-Aware Selection Large Language Model Training Bounded-Memory Peak Memory Efficiency

Jun 9, 2026cs.CL

REAL: A Reasoning-Enhanced Graph Framework for Long-Term Memory Management of LLMs

Large Language Models (LLMs) are increasingly expected to interact with users over long time horizons. However, due to their finite context window, LLMs cannot retain all past interactions, making long-term memory management essential for storing, updating, and retrieving historical information beyond the context limit. Although recent memory systems attempt to address this issue by storing historical information externally, existing approaches suffer from three key limitations: flat text-based memory organizations fail to capture explicit relations among memories, structured memory systems often destructively overwrite evolving facts, and current retrieval mechanisms remain query-agnostic and passive when evidence is incomplete. REAL constructs long-term conversational memory as a temporal and confidence-aware directed property graph, where each atomic fact is represented with entities, relations, valid-time intervals, confidence scores, and exploration intent labels. During memory construction, REAL adopts a non-destructive temporal update strategy that preserves parallel fact versions and their validity intervals, enabling faithful tracking of fact evolution. During retrieval, REAL anchors query-relevant root entities, decouples their exploration intents, and performs semantic evaluator-guided hybrid beam search to extract compact memory subgraphs. It further incorporates counterfactual inference to repair unreliable retrieval states and recover missing memory evidence through implicit logical relations. Comprehensive experiments demonstrate that REAL substantially improves long-term memory performance over flat-text, graph-based, and existing memory baselines, achieving an average improvement of 22.72%.

Keer Lu, Liwei Chen, Guoqing Jiang +3

School of Computer Science, Peking University · Kuaishou Technology · Center for Data Science, Academy for Advanced Interdisciplinary Studies, Peking University

Conversational Memory Peak Memory