Data Compression Methods

Follow this topic Ask about this topic

Momentum

46 papers in the last four weeks, up 39% on the four weeks before. 0.6% of all new papers.

Jul 6Week of Sep 21

Latest papers 324

Oct 1, 2026cs.SD

Q-SPT: Learnable Query-Based Compression for Low-Frame-Rate Speech Tokenization

Neural speech codecs increasingly serve as tokenizers for speech language models (SLMs). Lowering the frame rate reduces the computational and memory costs of SLMs, but makes it difficult to preserve both linguistic information and acoustic detail. Existing approaches rely on rule-based compression: average pooling can discard linguistic information, whereas similarity-based merging uses a fixed threshold on adjacent-frame similarity and applies the resulting boundaries to the acoustic stream. We propose Q-SPT, a low-frame-rate dual-stream speech tokenizer with separate, context-aware, learnable query-based compressors specialized for semantic and acoustic representations. In particular, queries at a fixed rate independently attend to the semantic and acoustic streams as separate key-value sources, enabling stream-specific, context-aware aggregation through two separately learned compressors. In addition, an autoregressive text loss explicitly supervises the semantic compressor to preserve linguistic information. Experimental results show that Q-SPT achieves the best reconstruction among the evaluated codecs at the same frame rate. In downstream SLMs, it yields the best speech recognition accuracy and text-to-speech perceptual quality with competitive intelligibility.

Jeeyoung Yun, Seohwan Yun, Sungwoong Kim

Korea University, Seoul, Republic of Korea

Neural Speech Codecs Audio Tokenizers Token Compression Speech Language Models Tokenizer Task-Aware Compression+3

Sep 30, 2026cs.CL

Sequential Functional Structured Tucker Compression for Large Language Model Attentions

Post-training compression of LLM attention is often formulated as independent matrix approximation, ignoring both the shared structure among attention projections and the representation shift introduced by earlier compression. We propose FTC, a sequential structured compression framework that adapts the approximation to the current compressed model while jointly exploiting the native Q/K/V head structure under a fixed storage budget. The output projection is handled separately to account for the changed post-attention representation. FTC requires neither fine-tuning nor gradient-based recovery. Across seven decoder-only LLMs from 6B to 32B parameters, FTC achieves the lowest WikiText-2 perplexity among the compared methods at every tested keep ratio on five modern GQA models, with the largest gains under aggressive compression. The improvements transfer to downstream tasks and remain substantial at the 32B scale.

Jiangfeng Chen, Xinyu Wang, Tianshuo Yan +4

University of Manitoba · McGill University · Simpleway +2

Large Language Model Compression Task-Aware Compression Data Compression Methods Dynamic Attention

Sep 30, 2026cs.LG

Learning Functional Subspaces for Neural Network Compression

Modern transformers pair impressive capabilities with substantial memory and compute demands. Low-rank weight factorization reduces both while keeping the matrices dense, and thus efficient on standard hardware. Existing methods, however, choose the subspace to remove from each weight matrix with local closed-form criteria: activation energy, layer-wise reconstruction error, or a quadratic approximation of the loss. These criteria ignore how errors propagate through the network, so at high compression the errors compound with depth and performance collapses. We introduce Learnable Subspace Projections (LSP), which instead learns the subspaces to discard end-to-end. Each linear layer, or tied group of layers that read the same activations, is assigned an orthogonal projector. All projectors are optimized jointly against a global objective--the KL divergence to the dense model's output distribution or the model's original training loss--while the pretrained weights remain frozen. Projectors are initialized from a whitened SVD truncation, and ranks are allocated by the output KL each projector induces per parameter saved. After training, the projectors merge into standard low-rank factors, with each tied group sharing one factor. In attention, this also lets the model cache one narrow latent in place of full keys and values. Across LLMs (OPT-125M/1.3B, Qwen3-4B, Llama-2-7B) and ViT-B/16, LSP outperforms baselines, and its advantage widens as compression increases. At -70% compression, LSP brings Llama-2-7B to 10.9 WikiText-2 perplexity and 42.2% mean zero-shot accuracy, versus 13.3 and 36.0% for the strongest baseline. The factorized model decodes up to 1.6x faster than the dense model at small batch sizes, and aching the shared latent shrinks the combined memory of weights and KV cache by 13.5x at a 128k-token context, versus at most 6.5x for untied baseline factorizations.

Massimo Bini, Anders Christensen, Stephan Alaniz +5

Helmholtz Munich · Technical University of Munich · MCML +6

Large Language Model Compression Subspace Low-Rank Structure Data Compression Methods Peak Memory

Sep 30, 2026math-ph

Cluster Attention Neural Operators for Solving Parametric Partial Differential Equations

Traditional simulations of parametric partial differential equations (PDEs) rely on repetitive computations for each parameter, which makes high-fidelity design impractical. Neural operators address this issue by learning solution operators, accelerating parameter-space mapping by orders of magnitude. Recent Transformer-based neural operators attempt to capture global dependencies, but often at the cost of quadratic attention complexity. Transolver resolves this problem by projecting physical states into a reduced slice space for attention computation. Although fast, this projection sacrifices fine spatial information. Moreover, by operating in this reduced space with shared weights across attention heads, it may constrain the model's flexibility, thereby limiting its capacity to capture complex phenomena. To address these issues, we propose the Cluster Attention Neural Operator (CANO), which reformulates attention via a novel cross-attention mechanism that dynamically clusters queries while preserving full-resolution keys and values. This avoids slice compression loss and removes weight-sharing limits. At the same time, the model remains fast without losing global interactions. Empirically, CANO achieves state-of-the-art performance across canonical PDE benchmarks, covering fluid and solid dynamics (e.g., Navier-Stokes, Airfoil, Plasticity), irregular unstructured geometries (e.g., Pipe Turbulence, Composites), and long-term temporal rollouts. Across solid deformation and turbulent flow benchmarks, CANO achieves lower errors than baselines and exhibits strong geometric adaptability and temporal consistency.

Ming Zhong, Antonio Colanera, Gianluigi Rozza +1

School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences, Beijing 100049, China. · State Key Laboratory of Mathematical Sciences, Academy of Mathematics and Systems Science, Chinese Academy of Sciences, Beijing 100190, China. · International School for Advanced Studies (SISSA), Trieste 34136, Italy. +1

Neural Operators Partial Differential Equations Data Compression Methods

Sep 30, 2026cs.LG

Beyond Uniform Compression: Budgeted Transmission Allocation for Extreme Federated Learning

Federated learning faces severe communication bottlenecks when clients upload high-dimensional model updates. Existing methods often compress these updates uniformly across all layers. This uniform approach ignores the heterogeneous value of different parameter blocks and wastes limited bandwidth on insensitive layers. To address this issue, we propose Layer-wise Budgeted Adaptive Transmission (LBAT). LBAT reframes federated communication under extreme uplink budgets as a resource allocation problem. Our framework dynamically estimates the transmission value of different layers utilising local training signals. It then employs an exact byte dynamic programming allocator to determine optimal rank and bit configurations under strict budgets. We validate LBAT on highly heterogeneous federated tabular prediction and data generation tasks. Extensive experiments demonstrate that LBAT consistently outperforms uniform rank, uniform quantisation, and fixed compression baselines across various extreme budget regimes. Furthermore, it achieves significantly better communication and utility tradeoffs while preserving essential distributional fidelity.

Pengfei Li, Mohammad Khalil

Centre for the Science of Learning & Technology (SLATE), University of Bergen, Bergen, Norway

Federated Learning Power Allocation Token Budget Allocation Data Compression Methods Distributional Fidelity

Sep 30, 2026cs.CV

DeCoPrune: Efficient KV-Cache Pruning for Autoregressive Video Diffusion via Denoising Consistency

Autoregressive video diffusion supports streaming generation and interactive control, but its KV cache grows with the generated history. Existing compression strategies discard history using fixed windows or select tokens through local attention and similarity signals, without directly measuring whether a chunk contributes information beyond the retained context. We introduce DeCoPrune, a training-free method that treats cache compression as a denoising-consistency problem. We find that tokens with larger discrepancies between intermediate clean predictions and final denoised values tend to carry visual evidence less predictable from the retained context. DeCoPrune uses this model-intrinsic signal to retain high-discrepancy tokens in the long-term cache while pruning low-discrepancy tokens. To evaluate information retention, we introduce CMBench, comprising 58 approximately one-minute generated or real-world context episodes and 116 Reappear or Revisit continuation tasks requiring recall of earlier events or objects. Experiments with LingBot World v2 show that DeCoPrune achieves a DINO score of 0.6701 on a 0-1 scale, with an 85.43% reduction in cumulative historical KV token counts and a 4.14-fold continuation-generation speedup over FullKV. Its head-specialized variant reaches 0.6783 at an 86.19% pruning ratio, approaching FullKV's 0.6803 score and exceeding the evaluated compression baselines at similar budgets. These results indicate that denoising consistency can support long-range information retention while reducing autoregressive inference cost. Our project homepage is https://decoprune.github.io. The code is available at https://github.com/DeCoPrune/CMBench, and the benchmark at https://huggingface.co/datasets/Aoraku/CMBench.

Zeqi Xiao, Qingle Liu, Kaiwen Zhang +3

Nanyang Technological University · Tsinghua University · Princeton University

Autoregressive Video Diffusion Models Iterative Denoising Methods Data Compression Methods

Sep 29, 2026cs.LG

LeanPolish: Verified Supervision for Lean Proof Compression

Verified proof edits offer a natural source of supervision for improving language-model-generated Lean proofs. Yet verification establishes that an edit is correct, not that its training signal is free of search artifacts. We introduce LeanPolish, a symbolic Lean 4 pipeline that releases 33,402 accepted local edits and 65,596 same-state failed attempts, and use it to study what models learn from this supervision. First-success search admits a goal-independent rule with perfect ranking accuracy; teacher-selected evaluation sites also reward trivial deletions. Continuing menu evaluation beyond the first success removes the ordering shortcut: a trained ranker selects the best candidate on 70.1% of evaluated held-out states, versus 36.9% for the strongest frozen baseline. For compression, iterating the symbolic pass raises miniF2F savings from 19.7% to 27.5%, exceeding the neural hybrids we test there. Verified neural editing helps on other proof sources, but matched frozen-model controls show that its gains need not come from training. The supervision does improve whole-proof rewriting: fine-tuning raises verified token reduction from 2.8% to 5.5% on 19 PutnamBench proofs. Together, the released edits, complete candidate pools, and controlled evaluations separate learning to imitate a search policy from improving on that search. They provide a reproducible basis for studying proof improvement while keeping correctness, compression, and edit policy distinct.

Pauline Bourigault

Imperial College London

Theorem Proving Proof Edit Data Compression Methods

Sep 29, 2026cs.AI

KV-Kaizen: Learning Context-Adaptive Cache Compression Choices

As the context size of text processed with an LLM grows, the size of KV caches can outstrip the memory allocated for the original model weights. This impacts LLM throughput negatively, since decoding is memory-bound and decode cost grows with cache size. Recent work alleviates this bottleneck by discarding the least relevant tokens. Eviction introduces a tension, since a one-off decision to discard content may prove detrimental later. Instead, we focus on alternative choices that can lead to cache compression without evicting tokens. We achieve this by learning a selector that is able to produce, based on context, a per-layer cache configuration towards an overall compression budget. The selector operates along three axes: sharing one cache across layers (depth), caching at fewer bits (precision), or truncating the low-rank latent cache representations (rank). We call the resulting method KV-Kaizen, for the many small per-layer choices it compounds. We observe that these interventions taken independently and uniformly over all layers limit achievable compression because they degrade accuracy. Crucially, composing them locally and adaptively to the context can instead preserve accuracy while achieving large memory savings. At inference, the selector runs once, before pre-fill. In evaluations on instruction following and reasoning tasks, our selectors reach the Pareto frontier of accuracy against cache size, against learning-free and post-hoc baselines. On long-context tasks, KV-Kaizen improves on eviction and can be composed with it, reaching a 32x smaller decode-time cache on a 14B model while preserving accuracy. A 4x cache size reduction incurs no accuracy degradation from 7B parameters up, and a compressed model is more accurate than a smaller uncompressed one with the same cache size. Together, these findings support pre-training large models and compressing them only afterwards.

Joao Monteiro, Louis Béthune, Anastasiia Filippova +3

Apple

Key-Value Cache Compression Large Language Model Compression Compressed Model Cache Data Compression Methods Long-Context Reasoning+1

Sep 29, 2026cs.AI

FOCUS: Training-Free Decision-Preserving Context Compression for LLM Agents

LLM agents accumulate interaction histories that grow linearly with task length, causing quadratic inference cost scaling and performance degradation from attention dilution. Existing context-compression methods learn what to discard offline: by contrastively optimizing guidelines, distilling compressors, or training compression policies. This incurs a substantial cost. Further, the compression policy is learned a priori and is not dynamically conditioned on the evolving test-time trajectories. In this paper we ask a complementary question: Which past interactions causally shape the agent's future decisions? We recast context compression as a causal decision preservation problem over discrete interaction units and introduce FOCUS, a training-free context compression framework that operates entirely at test time. Our method requires no offline data collection or fine-tuning, and is architecture-agnostic, attaching to any closed-API frontier model as a modular compression layer. We evaluate FOCUS on diverse agentic benchmarks including API and tool-calling, QA, web domain and multi-turn dialogue. Our method establishes new state of the art performance, cutting peak context by up to 48% and dependency by 73% while improving task success by up to 8.9 percentage points over uncompressed execution.

Shantanu Dixit, Anson Bastos, Xuchao Zhang +2

M365 Research, Microsoft

Large Language Model Agents Interaction History Agentic Benchmarks Data Compression Methods Training-Free

Sep 29, 2026cs.LG

Beyond Compression: Diagnosing How Post-Training Changes Mathematical Reasoning

Post-training is central to mathematical reasoning in modern large language models (LLMs), but endpoint pass@1 alone underidentifies what has changed. Gains may reflect newly reachable solutions, cheaper sampling of latent solutions, surface robustness, or memorisation. We compare three post-training paths under a common diagnostic readout: our sufficiently trained off-policy distillation trajectories, released Qwen3 off-policy-plus-on-policy distillation endpoints, and a released DeepSeek-Math endpoint trained with Group Relative Policy Optimisation (GRPO). Our probe uses cross-surface pass@K over verbatim prompts, paraphrases, numerical isomorphisms, and translations, plus consistency, distribution-shape, and verified supervised-fine-tuning (SFT) membership analyses. We find two regimes. On easier AMC problems, large-K ceilings are near saturation, so post-training mainly compresses sample cost. On harder AIME problems, post-training expands the large-K ceiling over the base model: sufficient off-policy distillation already raises this ceiling, Qwen3 released endpoints raise it further, and DeepSeek-Math GRPO does not dominate sufficient off-policy distillation at large K. English-dominant distillation improves non-English reasoning but preserves language-tier gaps. A controlled-overfit audit finds limited sensitivity in current SFT-membership probes. Compression is one regime of post-training, not a universal explanation.

Hongyang Li, Yiming Zhu, Xiao Li +3

University of Luxembourg · Seafill Open-Source Community · Université Paris-Saclay

Mathematical Reasoning Post-Training Math Problems Data Compression Methods

Sep 29, 2026cs.LG

Adapting Context Compression for Long-Horizon Agents with Counterfactual Continuations

Long-horizon agents require context compression to manage growing interaction histories. Compression quality, however, is ultimately determined by downstream execution. Existing prompt-adaptation methods infer compression errors by comparing full-context and compressed trajectories. Such comparisons cannot isolate individual compressions and are confounded by agent stochasticity. We first find that compression degrades reliability before solvability. Using matched counterfactual continuations that compare execution from the same agent state with versus without compression, we further show that severe degradation concentrates at isolated compression events. Motivated by this finding, we propose PAIR (Prompt Adaptation using Interventional Rollouts) for adapting structured compression prompts. PAIR identifies individual compressions that degrade subsequent execution, diagnoses their effects, and revises the relevant sections of a fixed compression template. PAIR achieves the strongest cross-run reliability among compressed methods in every main benchmark-scope combination, consistently exceeding the competing prompt-adaptation baseline. Without modifying the downstream agent, PAIR brings compressed execution close to the no-compression baseline and sometimes numerically exceeds it.

Guanghui Min, Liang Wu, Mingjia Shi +4

University of Virginia · Nokia

Long-Horizon Agents Interaction History Autoregressive Rollout Data Compression Methods Stochastic

Sep 28, 2026cs.LG

Periodic Weak Spots: Phase Sensitivity from Chunked KV-Cache Compression

Chunked KV-cache compression reduces the memory and attention costs of long-context inference by compressing windows of consecutive tokens into fewer cache entries at a fixed stride. Such compression also introduces a new positional coordinate: a token's phase, or its position relative to compression-window boundaries. We uncover a systematic asymmetry in models using such compression: the same information can be easy to retrieve at one phase and difficult at another. We call this periodic variation in retrieval performance phase sensitivity. In large open-weight models with such compression, long-context retrieval accuracy can differ by up to 40 percentage points across phases, revealing periodic weak spots that average benchmark scores can conceal. To investigate this behavior, we pretrain a family of transformers from scratch across multiple KV-compression designs, reproducing phase sensitivity across the variants. Mechanistic analysis using causal interventions in these models reveals phase specialization: different attention components contribute asymmetrically to retrieving information at different source phases. We further analyze idealized retrieval models, showing how gradient flow dynamics may favor sharp phase specialization. Evaluating models with chunked KV-cache compression thus requires measuring across compression phases: high average accuracy can coexist with systematic positional failures.

Xingyu Zhu, Pu, Yi +6

Luke · ByteDance Seed · University of California, Berkeley

Key-Value Cache Compression Data Compression Methods

Sep 28, 2026cs.LG

Cartridges++: KV Cache Compression without Off-Context Derailment

Serving long documents to a Large Language Model (LLM) repeatedly is expensive: computations grow with context length, and the memory footprint of the key-value (KV) cache balloons. Compressed KV (CKV) representations aim to mimic the cache of a document and are typically computed once and for all, ahead of inference time. Methods to obtain CKVs range from drop mechanisms that reduce their number of columns, to learned approaches. Among the latter, Cartridges have emerged as a leading compression method, learning compact KV representations through distillation on relevant Q/A pairs. While existing evaluations focus primarily on whether Cartridges and other CKVs yield approximately similar responses to document-related, on-context queries, we investigate the crucial deployment question of whether they can handle off-context queries, something the native KV representation is particularly good at, thanks to the mechanics of attention. We observe a fundamental trade-off: while Cartridges perform better for on-context queries, heuristic-variants preserve better the original LLM's ability to operate off-context. We measure this through their capability to avoid context contamination in their response, retain general knowledge, and follow instructions. We propose Cartridges++, simple modifications to cartridges that retain off-context abilities at small or negligible cost. The router variant decides at inference time whether the query should use the learned long-context memory, while the data-mixing variant allocates a small fraction of training Q/As to queries outside the reference long document. Our study shows that assessing CKVs on document utility alone can mask substantial degradation in broader model capabilities, yet those issues can be fixed with benign changes to CKV inference or training.

Sonia Laguna, Joao Monteiro, Marco Cuturi +2

Apple

Key-Value Cache Compression Efficient Long-Context Inference Cache Data Compression Methods Queries

Sep 28, 2026cs.LG

The Hidden Ratio in Adam: Stable Structure, Compression, and Sign Dynamics

Adam is the default optimizer for training modern deep neural networks, yet its adaptive behavior remains poorly understood due to the complex interaction between its first- and second-moment exponential moving averages (EMAs). We study Adam in the tied-

β

regime, where the two EMA decay rates are equal, and show that its adaptive dynamics can be expressed through a transformed ratio with approximately scale-stable behavior. Empirically, this transformed ratio exhibits a stable, heavy-tailed distribution across tasks, model scales, and training stages, in contrast to the variability of raw moment magnitudes. This empirical stability has both practical and conceptual consequences. First, we derive a recurrence for the transformed ratio, yielding a reparameterization of Adam that replaces the second moment with a compressible state. Leveraging its stable distribution, we show that a fixed 4-bit codebook is sufficient in our experiments to store this state without auxiliary scaling, achieving performance competitive with full-precision Adam. Second, the transformed ratio view clarifies Adam's connection to sign-based methods: Adam reduces to sign-based momentum modulated by the transformed ratio, and replacing it with a constant recovers Signum as a limiting case. This perspective further provides a simple rule for transferring learning rates between the two methods. Together, these results suggest that tied-

β

Adam admits a simple and approximately stable ratio structure underlying its adaptive behavior and demonstrate its utility for both analysis and efficient implementation.

Yihe Zhou, Tongtian Zhu, Yingxiao Huo +4

Department of Computer Science, The University of Manchester, Manchester, UK · College of Computer Science and Technology, Zhejiang University, Hangzhou, China

Adam Momentum Stochastic Gradient Descent Ratio Data Compression Methods Sign

Sep 28, 2026cs.CV

Beyond Selection: Token Parameterization for Extreme Visual Token Compression

Visual-token compression is effective for improving the efficiency of vision-language models, but under extreme compression budgets, token pruning can break visual grounding while learned resamplers increase parameter count, attention cost, and training complexity. We revisit compression through a token parameterization lens, separating (i) basis transformation and structured truncation (retained subspace/compressibility) from (ii) coordinate organization (optimization and cross-modal alignment). This view yields two coupled objectives, compressibility and learnability, which we formalize as unified functionals. Guided by these objectives, we design Braco, a lightweight four-step coder that combines transform-basis truncation, input-independent basis-coordinate embeddings, budget-dependent orthogonal re-parameterization, and learned spatial residual tokens from lightweight pooling. Experiments show that Braco forms the favorable empirical accuracy-efficiency frontier under

23\times

--

64\times

compression and remains competitive at

144\times

, reaching 95.2% accuracy while reducing prefill FLOPs by 84.2%--86.7% relative to the uncompressed upper bound. Against prior methods, Braco matches or improves accuracy while achieving up to approximately 36% end-to-end speedup and using

16.6\times

/

78.8\times

lower compressor latency/FLOPs.

Rui Zhong, Yu Li, Zheyu Yan +1

Zhejiang University

Token Compression Data Compression Methods

Sep 28, 2026cs.CV

Transform-Aligned Learned Features for Lossy Point Cloud Attribute Compression

Transform-based methods provide an effective framework for point cloud attribute compression by representing attributes as transform coefficients. Introducing learned spatial context into this framework requires mapping spatial representations to the transform domain, but this known basis change is often left for the network to learn implicitly. We propose Transform-Aligned Learned Features (TALF) by applying the attribute transform to learned spatial representations, explicitly aligning them with the coding targets. Our analysis shows that the resulting features exactly represent the first-order prediction term of a smooth nonlinear model, with a bounded Taylor remainder. We integrate TALF into a transform-based attribute codec with explicit coefficient prediction and conditional residual entropy modeling under a unified coefficient-domain rate--distortion objective, while retaining explicit quantization-step control. Extensive experiments across three benchmark datasets and multiple transform bases demonstrate that TALF improves rate--distortion performance over conventional and learned baselines.

Yueru Chen, Pengpeng Yu, Dingquan Li +3

Pengcheng Laboratory, Shenzhen, China · Sun Yat-sen University, Shenzhen, China · Peking University, Shenzhen, China +1

Point Clouds Rate-Distortion Theory Data Compression Methods Equivariant Learning

Sep 27, 2026cs.AI

CoViST: Visual Token Compression via Composable States

Visual token compression lowers the inference cost of vision--language models by representing images with fewer tokens. However, most existing methods compress visual tokens to a reduced set, leaving the amount of visual evidence represented by each token and its original spatial context implicit. Therefore, the compressed representation does not explicitly encode how much visual information each representative carries or where it lies in the original image. This limitation arises even after a single reduction and becomes more pronounced when compression is repeated across decoder layers. To address this issue, we propose CoViST, a training-free framework that represents a compressed image as a composable visual state. Specifically, the state combines representative features with original positions, effective contribution weights, and reusable selection metadata. CoViST constructs this state through coverage-guided selection and conservation-based contribution composition, and explicitly incorporates its contribution and positional information into decoder attention. Each component of the state retains its interpretation under successive reductions, enabling the same formulation to support both fixed compression before prefill and progressive compression within the decoder. Experimental results on seven LLaVA-1.5-7B benchmarks show that CoViST-Fixed retains 99.9%, 99.5%, and 98.1% of uncompressed performance at 192, 128, and 64 tokens, respectively, and CoViST-Pro retains 99.8%, 99.9%, and 99.1% at the corresponding layer-average budgets, outperforming state-of-the-art methods under their respective budget settings. Code will be released publicly.

Qi Zhang, Xiandong Meng, Ronggang Wang +1

Peng Cheng Laboratory Shenzhen, China · Peking University Shenzhen Graduate School Peking University Beijing, China

Token Compression Data Compression Methods Inference Cost

Sep 27, 2026cs.LG

When to Evict, Not What to Keep: Draft-Guided Eviction for Training-Free KV-Cache Compression

Training-free KV-cache compression methods such as SnapKV, H2O, and PyramidKV evict tokens at the end of prefill, aiming to preserve the attention mass that future queries are expected to use -optimizing what to keep. We show that this objective fails in two distinct ways. (1) Compensation: restoring the evicted attention mass can recover the attention-level target without recovering task quality. (2) Selection: covering more of the true decode-query mass can hurt quality when the recovered mass is fragmented rather than concentrated in coherent spans. These failures share a common cause: eviction occurs before the queries that determine the answer trajectory exist. We propose Draft-Guided Eviction (DGE), which defers eviction until after drafting the first k=2 answer tokens using the full cache - just one decode step beyond prefill. Because the draft is generated from the answer's own prefix, no cache entries are discarded before this trajectory signal becomes available. The per-head cache budget remains unchanged, and DGE can be applied directly to SnapKV, PyramidKV, H2O, and StreamingLLM without modifying their eviction scores. Unlike extra-pass methods, DGE changes when eviction occurs rather than what cache entries are selected. Extensive experiments demonstrate that DGE outperforms prior methods at every evaluated budget on five of six instruct-tuned backbones, achieving 44.2 on LongBench, nearly matching FullKV at 44.3. The timing-only control DGE-W achieves the same score, demonstrating that the gain comes from when eviction occurs rather than what is selected - an effect we term trajectory anchoring.

Haeyong Kang, Chang D. Yoo

Duksung Women’s University · KAIST

Eviction Key-Value Cache Compression Bounded-Memory Data Compression Methods

Sep 27, 2026cs.CV

FeCoSplat: Feedback-Guided Compression for Feed-Forward 3D Gaussian Splatting

Feed-forward 3D Gaussian Splatting (3DGS) enables efficient novel-view synthesis from sparse multi-view images, yet its representations remain costly to store and transmit. Existing approaches compress either the input images, incurring heavy receiver-side reconstruction, or the reconstructed Gaussian primitives, which are difficult to compress due to their heterogeneous and irregular attributes. We instead compress compact intermediate features, providing a better balance between compression efficiency and receiver-side complexity. Based on this paradigm, we propose FeCoSplat, a feedback-guided compression framework for feed-forward 3DGS. FeCoSplat first compresses multi-view features to obtain an intermediate 3DGS, whose rendered views are used as feedback to guide a second-stage compression for further refinement. The resulting bitstreams are decoded into a compact implicit state, from which the final Gaussian primitives are reconstructed with a lightweight predictor. Experiments demonstrate that FeCoSplat achieves favorable rate--distortion performance, particularly at low bitrates, while requiring only 3.45M parameters for receiver-side Gaussian reconstruction. Code will be released soon.

Yuxuan Li, Yihang Chen, Yufeng Zhang +2

Shanghai Jiao Tong University · Monash University

Feed-Forward 3D Gaussian Splatting Novel View Synthesis Data Compression Methods

Sep 24, 2026cs.CV

Towards Practical Compression of 3D Gaussian Splatting

3D Gaussian Splatting (3DGS) enables high-quality novel-view synthesis but requires substantial storage. Existing compression methods often rely on spatial context modeling over irregular 3D representations, increasing the complexity of training and coding. Meanwhile, floating-point context inference can introduce numerical inconsistencies across platforms, causing entropy-decoding failures. To address these practical challenges, we propose COSA-GS, which constructs context without spatial aggregation through anchor-wise causal factorization. Specifically, we use geometry context derived from each anchor's coordinates to model a compact learnable anchor latent. The anchor latent is then fused with the geometry context to form an anchor context for attribute coding. The resulting context model features a simple architecture composed solely of linear transformations and activations. We train COSA-GS using rate--distortion optimization with adaptive Gaussian pruning. Further, we develop quantization-aware training and integer inference for the context model to achieve bit-exact consistency of entropy-decoded symbols across platforms. Experiments demonstrate that COSA-GS achieves state-of-the-art compression performance while retaining fast and consistent cross-platform decoding, providing a simple yet effective framework for practical 3DGS compression. Code is available at https://github.com/pengpeng-yu/COSA-GS.

Pengpeng Yu, Yueru Chen, Fei Song +4

Sun Yat-sen University, China · Pengcheng Laboratory, China · Academy of Broadcasting Science, National Radio and Television Administration, China +1

Large-Scale 3D Editing Dataset Novel View Synthesis Data Compression Methods Video Latents

Sep 24, 2026cs.LG

Beyond Compression: Training Latent Representations for Stable Long-Horizon Rollout in Neural Surrogate Solvers

Latent neural surrogate solvers, or latent dynamics models, accelerate simulations of time-dependent physical systems by evolving a compressed latent space rather than resolving full-resolution fields directly. In principle this reduces computational cost and simplifies learning, but in practice errors often accumulate rapidly during long autoregressive rollouts, limiting predictive utility. We show that this instability does not stem from the latent representation itself, but arises when it is trained solely for reconstruction, producing representations poorly suited to long-horizon forecasting. We systematically evaluate training-level interventions that align latent representations with long-horizon rollout: Koopman operator learning and Hamming noise injection during autoencoder training to improve compression, together with noise injection and multi-step rollout fine-tuning to improve dynamics. Interventions that improve long-horizon rollout stability often degrade conventional training metrics, including reconstruction and one-step prediction accuracy. Collectively, these interventions reduce long-rollout error by approximately 40% and match or exceed the accuracy of full-resolution models on two physics benchmarks, while requiring 2 orders of magnitude fewer floating point operations and half the GPU memory. Applied to mesoscale crystal-plasticity simulations of high-cycle fatigue, the resulting surrogate achieves stable extrapolation over horizons orders of magnitude beyond those observed during training. More broadly, these results show that neural compression should be designed not merely to reduce dimensionality, but to restructure the solution space for stable dynamical evolution, a key requirement for reliable, efficient neural surrogates in scientific applications.

Andreas E. Robertson, Ashley T. Lenau, John D. Shimanek +5

Center for Integrated Nanotechnologies, Sandia National Laboratories, New Mexico, USA · Center for Computing Research, Sandia National Laboratories, Albuquerque, NM, USA · School of Engineering, Brown University, Providence, RI, USA +2

Latent Dynamics Future Latent Representations Surrogate Models Data Compression Methods

Sep 24, 2026cs.CL

MILO: Efficient Many-shot In-Context Learning with Block-wise Low-rank Compression

Many-shot in-context learning (ICL) enables large language models (LLMs) to adapt to complex tasks by conditioning on thousands of demonstration examples, but this paradigm shifts the inference efficiency bottleneck to the key-value (KV) cache memory. Due to the linear scaling behavior of the KV cache, storing these intermediate tensors has become a paramount challenge for both online serving and on-device deployment. To address this issue, we propose a novel compression framework, termed MILO, that exploits the low-rank redundancy inherent in many-shot contexts. Specifically, MILO features a block-wise low-rank compression strategy that compresses the KV cache at the block granularity, where each block contains multiple many-shot examples. Furthermore, to handle the heterogeneous context density across different blocks, MILO dynamically allocates rank budgets based on the information entropy, preserving the fidelity of critical blocks while aggressively compressing redundant ones. Experimental results on Qwen2.5 models demonstrate that our method achieves up to 50% reduction in KV cache memory and 1.8x throughput improvement, with negligible performance degradation on classification and reasoning benchmarks, significantly outperforming prior baselines.

Youpeng Zhao, Tian Tan, Liqian Peng +2

Google · University of Central Florida

Large Language Model Compression Key-Value Cache Compression LLM Inference Optimization In-Context Learning Low-Rank Structure Data Compression Methods

Sep 23, 2026cs.CL

Uncheatable Eval: Dynamic Compression-Based Evaluation of Language Models

Modern large language models are pretrained on massive datasets, making it difficult to prevent benchmark data from entering their training sets and undermining the reliability of evaluation results. Reliable evaluation is particularly challenging for base models, whose limited instruction-following ability complicates task-based assessment. We introduce Uncheatable Eval, a dynamic benchmark that regularly collects newly published text to evaluate base language models and reduce the risk of data contamination. Drawing on the relationship between a model's predictive ability and its ability to compress data losslessly, we use compression rate to evaluate how well models predict new text. We evaluate 80 models across 14 text categories, study how compression changes with context length, and examine the correlation between compression rate and zero-shot MMLU accuracy. Our results yield three main findings: (1) compression performance follows a consistent scaling trend with model size; (2) attention-based, hybrid, and recurrent models differ in how their compression performance changes as more context becomes available; and (3) lower compression rates are strongly associated with higher zero-shot MMLU accuracy. Code is available at https://github.com/Jellyfish042/uncheatable_eval.

Kaifeng Tan, Yudong Li, Linlin Shen

Shenzhen University · Tsinghua University

Context Length Data Compression Methods

Sep 23, 2026cs.CL

MORSE: Multi-Context Ordering via Reverse Scoring for Evidence-Preserving Compression

Retrieval-augmented generation often relies on multiple retrieved contexts that contain substantial redundancy, motivating context compression to preserve useful information under limited input budgets. Likelihood-based compressors can account for cross-context redundancy through sequential scoring, but this makes evidence scores dependent on context order. We show that permuting the same contexts under an unchanged compressor can substantially change which supporting evidence survives compression. We attribute this sensitivity to information preemption: earlier, partially relevant contexts can absorb credit for shared information, reducing the incremental scores of later, stronger evidence and increasing its risk of removal. Controlled pair-swap interventions provide direct empirical support for this mechanism by showing that placing stronger evidence before overlapping, partially relevant contexts can improve its survival. Based on this insight, we introduce MORSE, a compression-aware method for evidence-preserving context ordering. MORSE uses reverse query likelihood to construct an evidence-first anchor and to evaluate compressed candidate outputs, enabling compression-aware selection among alternative permutations. Across multi-hop Question Answering (QA) benchmarks, compression procedures, budgets, and scoring models, MORSE improves evidence retention over reverse ordering and generally outperforms matched random search, with downstream QA gains. Our code is available at https://github.com/tbn5pj/MORSE_code

Ke Wan, Yifan Wang, Liheng Lai +1

Department of Computer Science University of Virginia Charlottesville, VA, USA · AfterQuery

Data Compression Methods Bounded-Memory Scoring

Sep 22, 2026cs.AI

Optimizing the Score, Losing Sight of the Task: Reward Hacking Across Weights, Selection, and Prompts

A higher evaluation score does not always mean a better language model system. When optimization exploits an evaluator's mistakes, measured progress can conceal unchanged or deteriorating task performance. This failure can arise through parameter updates, selection among generated outputs, or revisions to persistent prompts. We develop a comparative framework for reward hacking across these three optimization substrates: weights, selection, and text. Building on the Proxy Compression Hypothesis and research on inference-time and in-context reward hacking, we examine how reachable behavior, optimization budgets, and persistent adaptation shape exposure to proxy error. We formalize a distance-dependent upper bound on evaluator disagreement and a capacity ordering for nested policy classes, then show why distance alone cannot establish a universal ranking of vulnerability. An exact finite-output illustration demonstrates how the location of a scoring defect changes the behavior favored by each method. We also map representative defenses across substrates, identifying which mechanisms transfer directly and which offer only functional analogies. Persistent prompts receive particular attention: their contents are inspectable, but the behavior induced by a small textual change may be difficult to anticipate. The formal analysis, numerical illustration, and published evidence together provide a basis for comparing optimization methods and identifying the conditions under which their defenses transfer. The resulting framework connects optimization choices to verification requirements: reliable improvement depends on controlling accessible failure modes and preserving evidence of task quality independent of the score being optimized.

Vansh Wahi

Destination-Only Scoring Scores Prompt Engineering Disagreement Data Compression Methods

Sep 21, 2026cs.AI

Learned Enterprise Data Comprehension: Compression and Routing for Data Agents

Structured-data agents in enterprise settings must reason over complex data environments whose relevant evidence is distributed across schemas, relationships, policies, and recurring business roles. Modern agentic systems often address this burden through reusable markdown-style memory or skill files that preserve previously discovered information for later queries, reducing the need to rediscover the same structure repeatedly. This is useful, but it obscures a natural division of labor: agents are well suited to semantic reasoning, while learned systems are well suited to predicting and organizing recurring structure. We introduce latent equivalence learning to bridge this gap. The framework separates persistent task-relevant identities from their dataset-relative realizations. In our realization, supporting and opposing evidence shape support-realized Gaussian prototypes that learn how those identities are expressed in a particular data environment, while soft-membership profiles retain distinctions lost under a hard assignment. A separate learned query-prototype system represents recurring evidential requirements and maps them through a learned compatibility function into the same persistent identity structure. This identity-factorized, query-conditioned routing materializes the relevant dataset-specific evidence for downstream reasoning, allowing the agent to operate over an already organized evidential state rather than reconstructing cross-schema structure at every query. On the Data Agent Benchmark, spanning 54 queries across 12 heterogeneous datasets, our full implementation achieves 94.67% dataset-macro stratified Pass@1 over five complete trials and 258/270 successful raw query attempts, compared with 55.51% for the benchmark's Claude Opus 4.6 reference agent, ranking first among 40 leaderboard entries at submission.

Ethan Torres, Eric Mills

Data Science Agents Data Compression Methods

Sep 20, 2026cs.CV

Beyond UV Mapping: Mesh Texture Compression via Surface-Aligned Texture Fields

Mesh texture compression typically relies on 2D UV atlases, whose chart discontinuities and mapping overhead can limit coding efficiency. To tackle this challenge, we introduce TexF, a surface-aligned texture field that organizes texture attributes in sparse voxels derived from the mesh surface. This representation supports high-resolution textures while preserving local 3D correlations for compression and enabling direct surface queries. For bitstream compression, TexF reuses established 3D attribute codecs, with voxel locations reconstructed from the decoded mesh without separate transmission. For GPU-resident compression, we develop 3DNTC, which combines quantized hash features with a lightweight decoder for random-access reconstruction at surface positions. Differentiable rendering enables image-space refinement of both voxel attributes and compressed neural fields. Experiments on the MPEG and AOM mesh compression benchmarks demonstrate improved average rate-distortion performance over representative UV-based methods for both bitstream and GPU-resident compression. 3DNTC also supports real-time rendering.

Jianqiang Wang, Junhui Hou, Siyu Ren +2

Department of Computer Science, City University of Hong Kong, Hong Kong SAR, China · Department of Electrical Engineering, Shanghai Jiao Tong University, Shanghai, China · Department of Computer Science and Engineering, Texas A&M University, TX 77840, US

Texture Meshes Voxel Data Compression Methods Extreme Ultraviolet

Sep 17, 2026cs.LG

Correct Now, Insufficient Later: Auditing Update Sufficiency in Context Compression

A memory can answer a current query correctly while discarding distinctions required by a later update. We investigate this failure with a paired-history audit: two histories have the same current answer, receive a shared future update, and require different subsequent answers. A pilot evaluates 24 history pairs across six synthetic mechanisms, 12 memory conditions, two repeats, and two model backends. A deterministic frontier selector obtains strict reveal accuracy of 96/96 on DeepSeek and 82/96 on GLM; a structured writer obtains 62 successes with one unresolved outcome and 56/96. The configured four-outcome joint contrast has finite-sample identification intervals of [0.521, 0.542] and [0.292, 0.313], not confidence intervals. A record-level audit distinguishes retained-state adequacy, response delivery, and answer-schema compliance without changing those original scores. It finds 26 and 25 well-formed but semantically wrong structured reveal memories, while all 14 GLM frontier reveal failures contain correct values in the wrong wrapper. Tombstone removal produces 16/16 exact replay failures in the targeted mechanism. Identifier renaming then exposes a separate flaw: original frontier late-reference adequacy falls from 8/8 to 94/320 transformed instances. We provide and test a label-equivariant repair, but it preserves only 2/8 original late-reference answers: eliminating a naming shortcut does not solve unknown future relevance. These results support a scoped evaluation methodology and reproducible failure analysis, not general superiority of the repaired algorithm. Paid pilot evidence, retrospective diagnostics, and new offline tests are reported separately; no independent held-out or natural-task validation is claimed.

Guangzhe Zhang

Independent AI Researcher

Model Auditing Correct Incomplete Answer Data Compression Methods

Sep 17, 2026cs.CL

DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression

The widespread adoption of long-horizon agents has made model workloads increasingly input-heavy. Although prior work has substantially reduced the cost of long-context computation, prefill remains computationally expensive, and large KV caches continue to strain HBM and SSD capacity and data-transfer bandwidth. Together, these compute, storage, and bandwidth demands constitute the primary bottleneck to further lowering deployment costs. To address this challenge, we introduce DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts (MoE) model with 552B backbone parameters and support for contexts of up to one million tokens. With its Causal Encoder-Decoder (CED) architecture, the model activates 16B parameters per token during decode but only 8B parameters during prefill, substantially improving cost efficiency for agentic workloads. To push the limits of KV cache compression, DeepSeek-V4.1-Flash combines cross-layer KV cache reuse in Compressed Sparse Attention 2 (CSA2) with FP4 KV caching. These designs reduce its global KV cache footprint (always in HBM) to 890 bytes per token, roughly 1/4 of the corresponding footprint of DeepSeek-V4-Flash. Further, through a dedicated deployment optimization known as SWA Bounded Replay, DeepSeek-V4.1-Flash reduces its persistent KV cache footprint (always on SSD or in host memory) to roughly 1/8 of that of DeepSeek-V4-Flash. Despite its much smaller KV cache footprint, the model delivers substantially better performance than the baseline. In addition, we streamline the DeepSeek-V4 architecture and introduce several efficient architectural extensions. We pretrain DeepSeek-V4.1-Flash on a multimodal corpus comprising 45T tokens and conduct comprehensive post-training, yielding strong performance across diverse text-based and multimodal agentic scenarios. Model checkpoints are available at https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash.

DeepSeek-AI, :, Anyi Xu +590

Key-Value Cache Compression Key-Value Cache Deepseek Block Sparse Flash Attention Long-Context Reasoning Data Compression Methods+1

Sep 17, 2026eess.IV

Compression Hurts, Pooling Helps: Information Loss in Rayleigh-Scale Estimation from B-Mode Ultrasound

Clinical B-mode images are widely available as potential data sources for quantitative ultrasound (QUS) analysis for tissue characterization. However, standard clinical ultrasound devices apply unknown log-compression to RF envelope data before display and storage. Previous work has demonstrated estimation of the underlying RF envelope statistics in the presence of an unknown compression law. Using Fisher information analysis, we show that finite-offset log compression causes severe information loss when estimating the Rayleigh scale

σ

, which controls diffuse speckle. For a single image window, unknown compression raises the minimum achievable variance for unbiased estimation of

σ

by a compression-independent factor of approximately

\FisherMinInflation

. When

M

equal-sized windows share the same unknown compression settings, the excess variance decays as

1/M

; even in the most favorable regime, reducing the variance inflation factor below

1.1

requires

\FisherBestCaseWindows

windows. Our analysis treats the contrast parameter

a

as unknown and the boundary offset

b

as known; estimating

b

experimentally shows even larger variance. We validate this theory using synthetic estimation experiments and demonstrate RF-scale recovery on real RF-envelope windows from the OASBUD dataset. Together, these results clarify the limitations of using routine B-mode images for QUS.

D. Hudson Smith, Ahmer Raza

Clemson University

Ultrasound Density Ratio Estimation Fisher Information Data Compression Methods Scaling Laws Global Average Pooling

Sep 16, 2026cs.CL

Beyond Outcomes: Dual-View Relational Learning for Efficient Agent Benchmarking

Agent benchmarks are substantially more costly to evaluate than conventional LLM benchmarks. Benchmark compression is therefore a natural solution, yet existing methods primarily model redundancy in task--model final-score distributions, which is important in agentic evaluation. To address this limitation, we analyze large-scale trajectories and identify six complementary process signals that are systematically associated with final agent performance. To disentangle agent performance redundancy from a complete perspective, we propose DualViewEval, an agent benchmark compression method that jointly exploits outcome and process relations to learn an exact-size miniset and predict the full-benchmark scores. Across five agent benchmarks and five representative baselines, DualViewEval achieves the best results in all datasets. With only 20 tasks, it achieves

24\times

--

40\times

compression on APEX-Agents and BFCL, reducing mean absolute error (MAE) by

14.5\%

--

28.2\%

over the strongest competitors while improving Kendall's

τ

by up to

7.2\%

relative to EssenceBench on SWE-bench Verified. The selected minisets further reveal capability differences among different agents, providing compact and diagnostic feedback for efficient agentic model development.

Xinshuai Guo, Junjie Wu, Dolly Deng +4

Hunyuan Team, Tencent · Tsinghua University

Agentic Benchmarks Large Language Model Benchmarks Data Compression Methods Relational Deep Learning

Sep 14, 2026cs.LG

Per-Matrix Optimality Is Not Enough: Three-Level Optimization for Low-Rank LLM Compression

Per-matrix singular value decomposition (SVD) truncation is Eckart-Young optimal in the whitened Frobenius norm, but errors from independently compressed matrices compound through the block's nonlinear forward pass. Inspired in part by hierarchical variational optimization in quantum many-body methods, we introduce a three-level chain that widens optimization scope from individual matrices to Transformer blocks to the full model: whitened SVD~(L1), block-level joint optimization~(L2), and end-to-end language-modeling loss refinement~(L3), all from 256 calibration sequences, with no instruction or recovery data. On LLaMA-7B at 60% compression, the chain reduces WikiText-2 perplexity from 42.1 to 19.1 to 11.4. The block-level stage acts as a regularizer: skipping it worsens Penn Treebank (PTB) perplexity by 24 points, a gap that additional end-to-end training did not close in our experiments. Perplexity gains hold across 20-80% compression, five architectures up to 13B parameters, and both in-distribution and out-of-distribution benchmarks, though the cross-architecture rows use architecture-specific configurations and the ratio sweep was not run under one common protocol. With more calibration data, skipping the block-level stage becomes competitive, revealing an offline compute--data trade-off. We therefore claim improvements only in perplexity and compression fidelity; downstream accuracy remains well below the dense model.

Huicheng Zhang, Xiyao Feng, Ze-Tong Li +6

Hong Kong University of Science and Technology (Guangzhou), Guangdong 511453, China · QudeLeap Research, Shanghai 200030, China

Large Language Model Compression Singular Value Decomposition Data Compression Methods Language Model Perplexity Optimality Fidelity+1

Sep 14, 2026eess.SP

Transforming harmonic coefficients for 3D splat compression

We address the problem of color attribute compression for 3D splats. We show that all images generated by 3D splats are linear in the coefficients for each color channel, each spherical harmonic, and each splat, and we identify a basis for the space of all such images. We identify an inner product for the coefficient space that induces the squared error loss on images. We show that orthonormalizing the coefficients with respect to this innner product before coding can yield over 2 dB gain.

Tam Thuc Do, Philip A. Chou, Gene Cheung

†York University, Canada · ⋆packet.media

Data Compression Methods Coefficients

Sep 14, 2026cs.CL

The Cost of Compression: A Rate-Distortion Limit on Factual Hallucination

Factual hallucination in closed-book question answering is often treated as a coverage problem: a model fails because the relevant fact is absent from its internal memory. This view misses a second source of error. Even when a fact has been observed, finite memory may force it to be stored only approximately. We study this effect through a simple coverage--compression model of factual recall. We consider an unstructured question-answering task with

N

possible queries and

K

possible answers. A learner observes

M

training facts, compresses them into at most

B

bits, and answers uniformly drawn test queries without retrieval. For a uniformly random ground-truth mapping, we prove

\mathcal{E} \geq \frac{M}{N}\delta^\star\!\left(\frac{B}{M}\right) + \left(1-\frac{M}{N}\right)\left(1-\frac{1}{K}\right)

, where

\delta^\star(r)

is the inverse rate-distortion function of a uniform

K

-ary source under zero-one loss. The two terms separate compression distortion on observed facts from missing coverage on unobserved facts. The bound gives a compact way to reason about selective memory, forced compression, structure, retrieval, abstention, and long-context organization. We study the predicted signatures with theory-implied simulations and controlled fact-injection probes in modern language models that vary fact load and effective trainable memory. The result is not a complete theory of hallucination, but an information-theoretic account of a separable failure mode: lossy recall of observed facts under finite memory.

Xi Wang, Shijia Xu, Rongfeng Guo

Hefei Institutes of Physical Science, Chinese Academy of Sciences, China · University of Science and Technology of China, China · Chongqing University, China +1

Factual Recall Visual Hallucinations Compressed Model Rate-Distortion Theory Data Compression Methods Question-Answer Pairs

Sep 14, 2026cs.LG

Scalable Discrete-to-Continuous Channel Simulation for Compression and Privacy

Channel simulation has recently emerged as a useful component in machine learning systems where samples from a prescribed probability distribution are to be compressed. Yet, general channel simulation algorithms often suffer from high computational costs, random stopping times or, in the worst case, can require generating an infinite number of shared random samples. We introduce a scheme for both exact and approximate simulation of discrete-to-continuous channels which conversely uses a fixed number of random samples, and therefore has a runtime independent of the channel and the input. Unlike existing channel simulation schemes which generate a sequence of independent samples from a proposal distribution, our approach generates one sample, or alternatively a fixed number of samples, from each potential target distribution. We then apply a latent permutation to the samples before performing sample selection using an exponential race. Our scheme provides a flexible tradeoff between the number of generated samples and the compression rate. Using polar and multilevel coding, we scale our approach to handle long blocklengths in

O(n \log n)

time in order to benefit from reduced per-symbol overhead. We conclude by demonstrating applications to variable-rate compression with stochastic VQ-VAEs and communication-efficient differentially private distributed mean estimation via exact simulation of the Gaussian mechanism.

Joseph Rowan, Buu Phan, Ashish J. Khisti

Department of Electrical and Computer Engineering, University of Toronto

Channel Vector-Quantized Variational Autoencoder Data Compression Methods

Sep 14, 2026cs.CV

Pixel Decodability Is Not a Compression Signal: Causally Evaluating Importance Proxies for Visual KV-Cache Eviction

Vision-language models retain a substantial amount of pixel-decodable visual content in their visual key-value cache. We show, in our setting, that this retention is task-inert: across our preregistered tests, how much a unit retains never positively tracks whether the computation that answers the question causally relies on it. We measure retention with a learned pixel-inversion decoder and causal use with single-super-patch KV ablation, the teacher-forced drop in gold-answer log-probability, and relate the two within images under a preregistered, sign-calibrated, held-out design. Retention is decoupled from attention and, in a well-powered null, from causal utilization. Utilization is not inert to every proxy: attention weakly but significantly tracks it, the only signal we find that does and the design's positive control. We characterize pixel-decodable retention as an informational axis of the visual KV cache, orthogonal to the functional one. How much task-inert content a cache holds differs by architecture in our model pair: the encoder-free model retains 2.7 times more than the encoder-based one. The engineering consequence is a controlled negative result. At super-patch granularity, deconfounded pixel-decodable retention ranks KV eviction no better than random; at token granularity it acquires only a weak inverse-importance signal at larger budgets, dominated at every budget by attention magnitude. In our setting, pixel-decodable reconstructability is not a competitive KV-compression signal at any granularity we test.

Chenyu Zhou, Qiliang Jiang, Shuning Wu +1

School of Engineering, Institute of Science Tokyo, Japan · College of Control Science and Engineering, Zhejiang University, China · Department of Electrical and Computer Engineering, National University of Singapore

Data Compression Methods Pixels Importance Causal

Sep 14, 2026cs.CL

ESTS at WMT26: Routing-Informed Expert Pruning for Model Compression

We describe six submissions under the team name ESTS to the unconstrained WMT26 Model Compression Shared Task for English--Simplified Chinese and English--Egyptian Arabic. We submit three compression operating points per translation direction, all derived from GPT-OSS-20B. We use task-specific routing mass to rank experts and cross-lingual routing divergence to allocate retained capacity across layers, then physically remove low-importance experts. The resulting specialists are recovery-tuned on GPT-5.1-generated synthetic translation data and further compressed by applying MXFP4 quantization to the retained expert projection weights. We additionally implement a robust inference system for the instruction-conditioned WMT26 setting, including category inference, output validation, retries, segmented fallback, and source-owned JSON reconstruction. Across our six submissions, parameter counts range from 4.186B to 7.770B and packed artifact sizes from 4.55 to 6.33~GiB. Internal xCOMET-XL evaluation using GPT-5.1 pseudo-references provides an internal comparison across the submitted compression operating points.

Liu O. Martin, Lucas Bandarkar, Nanyun Peng

University of California, Los Angeles

Large Language Model Compression Data Compression Methods Cross-Lingual Consistency

Sep 14, 2026cs.CL

Zipbench: Low-Cost Framework for Compressing Comprehensive Benchmarks of Large Language Models

Comprehensive benchmark suites are essential for improving large language models (LLMs), but many widely used benchmarks are redundant, making evaluation unnecessarily expensive. Although recent benchmark compression methods (BCMs) can mitigate this cost, many strong BCMs rely on large collections of per-sample evaluation results from numerous LLMs to identify representative samples. Building such collections is also expensive unless they are already public, making these methods difficult to extend to newly released benchmarks. To address this challenge, we present ZipBench, a simple and low-cost BCM with theoretical error and rank-consistency guarantees. ZipBench evaluates only a small set of anchor LLMs, synthesizes pseudo evaluation results to broaden coverage, learns compact sample representations, and selects a small yet representative subset. Building on it, we create ZipBench Zoo, a collection of compact versions of 100+ benchmark proxies spanning text, multimodal, and agent tasks. These benchmark achieve mean absolute errors of 0.002--0.02 and average Spearman correlations of ~0.98 with the full benchmarks. Overall, ZipBench reduces the cost of both LLM evaluation and compact benchmark construction, lowering the barrier to broad LLM research for compute-constrained researchers. The code has been released in https://github.com/MilkThink-Lab/ZipBench.

Zhongzhan Huang, Junxin Li, Guoming Ling +3

Bosch Research · Sun Yat-sen University

Large Language Model Benchmarks Large Language Model Evaluation Data Compression Methods

Sep 13, 2026cs.CL

Pull: Lazy Materialization of Working Memory for Stateful LLM Conversations

As LLM conversations grow to hundreds of turns, full-context injection incurs

O(N^2)

cumulative token costs, while lossy summarization or hard truncation irreversibly discards historical state. We propose Pull, a session router that maintains an addressable metadata directory via a local, deterministic Purifier (zero LLM calls, millisecond-level latency). At query time, the LLM lazily materializes only the turns it needs; unmaterialized turns remain accessible but collapsed. Unlike irreversible compression, Pull's materialization is reversible; subsequent queries can expand any collapsed turn. On LoCoEval (128 conversations, 12,780 turns), Pull reduces per-query context tokens (Phase 2) by 75.1 percent on single-hop tasks with equivalent quality (

Δ= -0.002

, n.s.) and by 72.0 percent on multi-hop tasks with no quality loss (

Δ= +0.017

). A controlled routing benchmark (7,831 queries x 10 methods) shows that entity lifecycle tracking is empirically a prerequisite for distance-independent routing. On BEAM 1M (14 conversations, 263 questions), Pull improves F1 by +55.2 percent over a truncation baseline.

Jiangang Chen

Chengdu Beiluoshimen Technology Co., Ltd.

Multi-Turn Large Language Model Conversations Working Memory Data Compression Methods

Sep 12, 2026cs.AI

Memory Compression for High-Fanout Agent Sandboxes

High-fanout agent workloads create a growing memory bottleneck because a single task may spawn many concurrent sandbox sessions. Yet these sandboxes are far from independent: they originate from a shared template and execute related trajectories, exposing substantial template-relative and cross-sandbox memory redundancy. Conventional memory compression is poorly matched to this setting in three fundamental dimensions: how to compress, because they fail to exploit similarity across non-identical sandbox pages; what to compress, because they control page-fault overhead through conservative page selection; and when to compress, because compression is either triggered by memory pressure or performed without awareness of agent execution phases. We present AgentZip, the first memory compression system designed specifically for AI-agent sandboxes. AgentZip introduces compression mechanisms that exploit both the template-relative and cross-sandbox redundancy. It broadens the compression scope to any page with a profitable representation and shifts overhead control from compression-time page selection to restore-time prefetching. It further aligns expensive compression with LLM waiting periods to avoid interfering with foreground tool execution. Across LLM training and inference workloads, AgentZip reduces sandbox-owned memory by up to 8.7x, compared with 2.1x for the Linux configuration. Restore prefetching and agent-execution-aware scheduling reduce the slowdown of aggressive compression from as high as 3.1x to 1.40x while retaining nearly all of its memory-saving benefit.

Mengming Li, Ceyu XU, Qijun Zhang +4

HKUST

Sandbox Task-Aware Compression Peak Memory Data Compression Methods Inference Workloads Cp-Agent

Sep 11, 2026cs.SD

Copying Versus Randomization in Lempel-Ziv Music Synthesis

We utilize Lempel-Ziv universal compression for music note generation. We control the algorithm's tendency to over-copy or under-copy training data by manipulating the average sequence length saved in the dictionary.

Nadav Mishan, Ram Zamir

Department of Electrical Engineering Tel Aviv University Tel Aviv, Israel

Synthesizer Data Compression Methods Random

Sep 11, 2026cs.CL

FlexComp: One Model for Every Ratio in Context Compression

Soft context compression condenses a context into a few memory tokens that a frozen LLM consumes in place of the raw text, but existing compressors fix the compression ratio at training and inference: each deployed ratio requires a separately trained model, and the chosen ratio is applied uniformly to all inputs, whose actual needs vary drastically. We propose FlexComp, a method-agnostic framework that decouples the ratio from both training and deployment: Matryoshka-style training samples the memory budget

K

per instance, turning one model into an any-ratio compressor, and the budget is then chosen per input by: (1) confidence-based cascade routing or (2) a lightweight learned

K

predictor. Across ICAE, 500xCompressor, and SAC on MRQA, a single FlexComp model matches separately trained fixed-ratio specialists with minimal degradation. Cascade routing preserves over 98% of the mildest ratio's accuracy at up to 266x average compression; the

K

predictor, in a single compression-decoding pass, reaches 158-236x within 0.7 F1 of the mildest ratio. At serving-scale batch sizes, the

K

predictor cuts context KV cache by 50% and improves decoding throughput by 47%.

Kaiyan Zhao, Zhongtao Miao, Akiko Aizawa +1

The University of Tokyo · National Institute of Informatics

Data Compression Methods Key-Value Cache Ratio

Sep 8, 2026cs.LG

Length Generalization for Transformers via Compression

Recent advancements in transformer length generalization theory enable us to reliably predict when a transformer can learn to solve a task. In particular, the C-RASP hypothesis (a formalized version of the so-called RASP-l conjecture) posits that transformers length-generalize on a task if and only if a solution is expressible in the C-RASP language. While this hypothesis has strong empirical validation, theoretical problems arise from the fact that no computable length generalization bounds exist for C-RASP, alongside the discovery of seemingly contradictory experiments. To address these problems, we refine the C-RASP hypothesis utilizing the recently-proposed fragments C-RASP+ and C-RASP1. These fragments have computable length generalization bounds, though in the worst case requiring an extremely large (double exponential) sample size. It is an open question whether these sample size bounds are tight. In this paper, we resolve this open question by providing an exponentially tighter bound. In doing so, we show a polynomial length generalization bound for transformers if we adopt compressed strings, via a novel connection to power words. As an application, we show how this yields a fine-grained analysis of the C-RASP conjecture that resolves contradicting experimental evidence against it.

Georg Zetzsche, Hongjian Jiang, Andy Yang +4

Max Planck Institute for Software Systems (MPI-SWS) · RPTU Kaiserslautern-Landau · University of Notre Dame

Transformer Architectures Generalization Bounds Data Compression Methods Open Problems Sample Size

Sep 8, 2026cs.SE

AttnCompress: Dynamic Attention-Guided Trajectory Compression for Software Engineering Agents

The transition from human-centric assistance to Autonomous Software Engineering (ASE) agents has enabled the resolution of complex real-world SE tasks. However, the trial-and-error nature of these agents generates lengthy interaction trajectories, creating severe bottlenecks in terms of context window limits and cost. While context compression offers a potential remedy, prior approaches suffer from static pruning strategies and granularity mismatches, often failing to preserve the semantic dependencies and syntactic details crucial for SE tasks. To strictly preserve critical task evidence while reducing context length, we introduce AttnCompress, a dynamic attention-guided trajectory compression framework. Unlike existing approaches, AttnCompress bridges the gap between semantic integrity and dynamic adaptability through three key mechanisms: (1) structure-aware segmentation via perplexity (PPL) spikes to preserve the syntactic structure of code and logs; (2) relevance estimation using proxy attention weights to quantify the precise relevance of historical blocks to the agent's current reasoning; and (3) a dynamic rolling window to re-evaluate and recall historical context as the task evolves. Extensive evaluation on SWE-Bench-Verified and Multi-SWE-Bench demonstrates that AttnCompress achieves a pass rate of 53.17%, outperforming prior state-of-the-art baselines while reducing token consumption by 21.6% and total costs by 33.6%. The framework proves to be model-agnostic and generalizes effectively across diverse programming languages.

Zhengran Zeng, Yixin Li, Rui Xie +2

Peking University, China

Task-Aware Compression Data Compression Methods Structured Pruning

Sep 7, 2026cs.LG

MetaKV: Adaptive KV Cache Compression for Constrained LLM Inference

Key--value (KV) cache compression is an effective way to reduce the memory overhead of large language model (LLM) inference, particularly for long-context workloads. However, existing compression methods make different trade-offs among accuracy, inference latency, and peak KV cache memory utilization, making a single fixed configuration unsuitable across different prompts and resource constraints. We introduce MetaKV, an adaptive framework that selects a KV cache compression configuration for each input prompt based on user-specified latency and peak memory budgets. MetaKV uses lightweight prediction models to estimate the end-to-end latency, peak memory, and probability of a correct response for each candidate configuration, and selects the configuration that best satisfies the latency-memory constraints while preserving accuracy. We evaluate MetaKV across ten configurations from three representative KV cache compression methods, KVQuant, H

_2

O, and RocketKV, together with an uncompressed FP16 configuration, on four datasets covering mathematics, science, commonsense reasoning, and reading comprehension. Across a wide range of latency and peak memory constraints, MetaKV consistently outperforms the best static configuration, improving constrained success rate (CSR), the fraction of prompts answered correctly while satisfying both constraints, by approximately 0.07 on average and up to 0.135. These results demonstrate the benefit of adapting KV cache compression to individual prompts and latency-memory constraints. Code is available at https://github.com/MichaelWang0505/MetaKV.git

Michael Wang, Keith Li, Roozbeh Bostandoost

Lake Washington School District · XSchool · UMass Amherst

Key-Value Cache Compression LLM Inference Optimization Data Compression Methods Peak Memory Inference Latency Meta

Sep 7, 2026cs.LG

Dense Structural Compression of Transformers via Gauge-Correct Channel Removal

Inference energy per token drives the cost and carbon footprint of deployed transformers. It is dominated by dense matrix products that incur fused multiply-accumulate (FMA) operations and memory traffic. To reduce these computations while retaining dense tensors for high GPU throughput, we develop a methodology from first principles to adapt structural complexity during training to maximize inference utility per unit compute. Channel penalties drive entire tensor slices to zero to enable physical removal while preserving density and the network function. The natural approach, penalizing the norm of operator components acting through each channel, is provably destabilized by gauge freedom. We resolve this pathology with GaugeLasso: additive symmetric group-lasso penalties that recover a monotone function of product-norms when the network converges to gauge balance. Our equilibrium analysis enables per-channel calibration to correctly suppress slices that under-perform in inference utility per unit compute. Under adaptive pressure, the network reorganizes into depth-dependent structural profiles that can be far smaller than the architecture required to learn the task. On polynomial long division over

\mathbb{F}_{31}

, compute compresses from 148 to 255 times with perfect accuracy. On character-level language modeling, compressed models outperform the hand-designed baseline at equal FMA. On masked autoencoding, a compression trial exposes which axes were over-provisioned and which saturated, guiding a better second design. Compaction also accelerates training monotonically as the model progresses. Post-hoc pruning with the same utility ranking cannot reach these structures, showing that sustained pressure is central to discovery of efficient models. Retraining a discovered architecture recovers baseline quality on our statistical tasks, but fails on our exact algorithmic task.

Jed A. Duersch, Naïm Es-Sebbani, Nathanaël Haas +1

Université d’Artois, CNRS, CRIL UMR 8188, Lens, France

Data Compression Methods Transformer Architectures Matrix Multiplication Channel Language Modeling

Sep 7, 2026cs.CV

Mind the Approximation: Fisher-Weighted SVD Compression for ViTs

Model compression is key to mitigate deployment challenges of ever growing machine learning models. In this area of research, singular value decomposition (SVD)-based compression offers a compelling trade-off between computational efficiency and model accuracy. Fisher-weighted SVD in particular provides principled, loss-aware compression. However, we find that improving the fidelity of Fisher approximation used in the compression is poorly predictive of post-compression accuracy for Vision Transformers (ViTs). Motivated by this observation, we propose FACTS, a structured Fisher Approximation tailored to Compressing ViTs with Fisher-weighted SVD, which enforces token-local aggregation while preserving within-token activation-gradient dependence. Additionally, we introduce a fast Constrained Rank Search (CoRS), that optimizes layer-wise rank allocation while adhering to a fixed floating point operation (FLOP) constraint. Extensive experiments across ViTs and hybrid architectures demonstrate that FACTS consistently improves accuracy-efficiency trade-offs without requiring finetuning. Notably, it outperforms the strongest SVD baseline by up to +5.8 percentage points (p.p.) Top-1 on Swin-B, with further gains driven by our search method. Code is available at https://github.com/MoritzTho/FACTS.

Moritz Thoma, Maximilian Groezinger, Maximilian Forstenhäusler +7

Technical University of Munich, DE · BMW Group, DE · University of Glasgow, GB +1

Vision Transformer Singular Value Decomposition Data Compression Methods Approximation

Sep 4, 2026cs.CL

Compression Beyond the Uncompressed: A Two-Stage Training Recipe for Soft Context Compression in RAG

Retrieval-Augmented Generation (RAG) improves knowledge-intensive generation by conditioning language models on retrieved documents, but processing these documents becomes increasingly expensive as retrieval depth grows. Soft context compression reduces this cost by encoding documents into compact continuous representations that can be precomputed and reused across queries. However, many existing methods train compressed models by distilling from a full-context teacher. When the teacher is wrong, such distillation can reinforce its errors, while teacher imitation provides no direct signal for improving beyond the teacher. We propose DEX-Comp, a two-stage training recipe that separates reliable imitation from targeted exploration. Pure Distillation learns only from teacher-correct questions to mitigate error propagation, while Hard Exploration applies outcome-based reinforcement learning to teacher-failed questions to directly optimize answer correctness. Across five open-domain QA benchmarks and retrieval depths from top-

5

to top-

30

, DEX-Comp at

16\times

compression outperforms all evaluated compression baselines and surpasses the untuned full-context RAG model in average accuracy, while reducing time-to-first-token by

4.4\times

--

23.7\times

. Evaluations across additional datasets and backbones further demonstrate its generalization.

Shuyu Guo, Shuo Zhang, Zhaochun Ren

Shandong University Qingdao, China · Bloomberg London, United Kingdom · Leiden University Leiden, The Netherlands

Search-Augmented Reasoning Data Compression Methods Multi-Stage Training Teacher

Sep 3, 2026cs.CV

Tree-Structured Vector Quantization For Efficient And Progressive Image Compression

Vector-quantization based image compression has achieved strong rate--distortion performance, yet most of them still produce a separate compressed representation for each target bitrate. Such variable-rate behavior allows one model to operate at multiple rates, but it does not necessarily provide a progressive bitstream whose prefixes are themselves decodable and can be refined by appending additional bits. We propose \textbf{Tree-VQ}, a progressive tree-structured vector quantization framework for learned image compression. Tree-VQ organizes discrete codewords as a hierarchical binary tree and represents each latent token by a routed root-to-leaf path. Crucially, every prefix of this path corresponds to a valid quantized representation, so shallow internal nodes serve as coarse reconstruction codes and deeper nodes provide successive refinements. This allows a compressed image to be decoded from an early prefix and progressively improved as more branch symbols are received, rather than being re-encoded for different target rates. To make this structure practical for compression, we introduce a prefix-compatible tree entropy model that codes progressive continuation decisions and routed branch refinements using only causally available decoded contexts. We further use rate-aware refinement scheduling to decide which spatial blocks should receive additional tree bits under a given prefix budget, and hierarchical prefix supervision to ensure that internal nodes are directly decodable at low rates. Experiments show that Tree-VQ achieves a superior performance--efficiency trade-off, delivering the best perceptual compression results with much fewer parameters and lower latency than competing methods.

Xinkun Wang, Tianyi Xu, Qingyu Luo +4

School of Artificial Intelligence Xidian University

Residual Vector Quantization Data Compression Methods Rate-Distortion Theory

Sep 3, 2026cs.LG

On the Interaction Between Model Compression and Test-Time Adaptation

Deep neural networks deployed in the wild must be both efficient and adaptable, requiring model compression and test-time adaptation (TTA). While both are well studied in isolation, their interaction remains poorly understood. We systematically analyze how structured compression affects a model's ability to adapt under distribution shift. Using ResNet-18 and ViT-Base on CIFAR-10-C and ImageNet-C, we evaluate multiple compression methods combined with standard TTA techniques. We introduce a diagnostic framework that examines representational expressivity and adaptation subspace compatibility. Our results reveal a consistent gap: although compressed models retain high accuracy under supervised adaptation, their TTA performance degrades significantly with increasing compression. We show that this stems from reduced representational diversity and structural constraints that limit recoverability. These effects strongly depend on the compression method, highlighting the need to design compression strategies that preserve adaptability.

Francesco Corti, Dong Wang, Young D. Kwon +2

Graz University of Technology Graz, Austria · Samsung AI Center-Cambridge University of Cambridge Cambridge, United Kingdom · University of Cambridge Cambridge, United Kingdom +1

Stable Test-Time Adaptation Compressed Model Test-Time Adaptation Data Compression Methods Subspace Expressivity

Sep 3, 2026cs.CV

Neural Video Compression Based on Deformable Temporal Alignment and Difference-aware Fusion

In conditional coding-based neural video compression, the quality of temporal context directly affects compression per- formance. Existing methods mostly construct context from prop- agated reference features, but they are vulnerable to motion esti- mation and local alignment errors in regions with complex mo- tion, occlusion, and high-frequency textures, resulting in inaccu- rate temporal information. To address this issue, this paper pro- poses a method combining deformable temporal alignment and difference-aware spatial selective fusion. A Context-aware Tem- poral Alignment Module is used to generate complementary tem- poral context, while a Difference-aware Spatial Selective Fusion module adaptively selects reliable temporal information and sup- presses misalignment. Experiments show that the proposed method achieves certain rate-distortion performance improve- ment over DCVC-DC.

Chuyue Shan, Songlin Sun, Wang Chenwei +1

Beijing University of Posts and Telecommunications Beijing, China · Google · Nanjing University

Video Compression Data Compression Methods Rate-Distortion Theory

Sep 2, 2026cs.AI

Post-Training Ternarization of Qwen3-4B Capability, Effective Bit Budget, Storage Compression, and Deployment

Ultra-low-bit language models can reduce storage and memory bandwidth, but a nominal "1.58-bit" label does not fully describe the stored representation, retained capability, or runtime behavior. We study an end-to-end post-training conversion of Qwen, an instruction-tuned 4B-parameter model, using KOTMS rotation, E2M-ATQ ternarization, and GPTQ-style error compensation from TWLA. The experiment is weight-only: activations remain at 16-bit precision, so ILA-AMP is omitted. We evaluate effective bit accounting, task capability retention, perplexity, calibration sensitivity, checkpoint composition, and deployment behavior. The final conversion uses 1.641 effective bits per weight for quantized linear weights, with 81.62% of model parameters targeted. Across ten scored capability comparisons, accuracy falls from 64.5% to 54.7%. Degradation is uneven: BoolQ retains 84.6% chance-corrected teacher performance, while ARC-Challenge retains 43.8%. Perplexity rises from 13.639 to 18.748 on WikiText-2, 24.700 to 31.992 on PTB, and 19.831 to 28.966 on C4. A subsequent packing run preserves the ternary planes and scales, reducing reported model size from 8.29 GiB to 3.96 GiB with essentially unchanged perplexity. A separate third-party packing attempt was lossy and is excluded from the primary artifact claim. The packed artifact has not been benchmarked end-to-end for task accuracy or generation throughput. A preliminary Triton GEMV microbenchmark is 4.6x slower than FP16 cuBLAS on one tested shape. We therefore do not claim that compression alone yields faster inference.

Anirudh Malik, M Sparsh Mehra, Poojith Devan

OneBit AI

Qwen3 Small Large Language Models Full-Precision Performance Data Compression Methods Post-Training Token Budget Allocation

Sep 1, 2026cs.LG

Emergence of Fibrations, Compression, and Symmetry Breaking in Artificial Neural Networks

Artificial neural networks are often regarded as powerful yet opaque black boxes. Here, we demonstrate that learning in deep neural networks generates local symmetries known in graph theory as fibrations and coverings. We prove that covering symmetries are stable attractors of stochastic gradient descent. Consistent with this theory, we report the emergence of covering symmetries across major network architectures, including multilayer, convolutional, recurrent, and transformer networks. Exploiting these symmetries enables drastic model compression - reducing networks to 17% of their original size without sacrificing performance. Furthermore, controlled breaking of covering symmetry overcomes the loss of plasticity, achieving state-of-the-art performance in continual learning. The theoretical results provide a new foundation for AI systems based on symmetries that convert black boxes into interpretable colored graphs and enable more efficient inference and lifelong learning.

Osvaldo M Velarde, Lucas C Parra, Alireza Hashemi +1

Levich Institute and Physics Department, City College of New York, 160 Convent Ave, NY, 10031, NY, USA. · Biomedical Engineering Department, City College of New York, 160 Convent Ave, NY, 610101, NY, USA.

Symmetry Compressed Model Attractors Data Compression Methods Emergence

Sep 1, 2026cs.LG

Linear Reusable Neural Bases Architecture for Network Compression

Memory constraints remain a critical bottleneck in the deployment of large-scale AI models. Parameter sharing across network depth reduces model storage, but repeatedly applying an identical transformation limits flexibility across layers. Inspired by time--memory trade-offs in classical algorithms, we introduce the Linear Reusable Neural Bases (LRNB) architecture, an RNN-based framework that improves parameter efficiency through parameter reuse at the \textit{neuron level}. Each feedforward residual module is represented as a linear combination of shared neural bases, with depth-specific learnable coefficients and optional shifts providing flexibility across layers. This formulation reduces parameter redundancy across depth and enables the construction of wider and deeper networks within a fixed parameter budget. We further provide a geometric interpretation of the neural bases from a vector-field perspective and extend the framework to linear projection modules. Experiments demonstrate that the LRNB architecture achieves comparable or lower final training loss than independently parameterized baselines while using fewer parameters and maintaining stable training dynamics. These findings support neuron-level reuse as a practical approach to parameter-efficient network design.

Binshuai Wang, Peng Wei, Mahyar Ghazanfari

Department of Computer Science The George Washington University Washington D.C., USA

Compressed Model Recurrent Neural Networks Data Compression Methods Peak Memory Artificial Intelligence Models Mathematics

Sep 1, 2026cs.LG

LatentPress: Context Compression Beyond Text and Vision

Compressed context is usually carried as human-readable text or as rendered images that must be decoded, even when its consumer is a language model. We introduce LatentPress, which writes conversational histories and long documents into a third representation: continuous memory tokens that a frozen decoder reads directly through its input-embedding interface, with no text reconstruction at inference. A small reader-matched writer compresses

4

-

16\times

while training only an adapter (4.2M-26.2M parameters,

\sim\!0.1\%

of the decoder). On LongMemEval, LatentPress reaches

0.504

accuracy at

7.70\times

compression versus

0.490

for uncompressed evidence, outperforming text summaries (0.184) and OCR-based compression (0.426 to 0.312). On LongBench-QA, in-domain writers match or exceed raw-context reading at

4

-

8\times

compression, while

16\times

trails raw. Writing takes 43ms per conversation, roughly an order of magnitude faster than text summarization or OCR reconstruction, and reading is

5

-

9\times

faster than raw context or cached OCR. We validate the interface under two transfer settings, zero-shot from UltraChat to LongMemEval memory QA and from LongMemEval-derived QA to unseen LongBench document domains, establishing direct soft tokens as a practical machine-facing context interface beyond text and vision. The implementation of the experiments could be found at: https://github.com/HJSang/LatentPress .

Zhengze Zhou, Hejian Sang

1Cornell University · 2Iowa State University

Efficient Long-Context Inference Large Language Model Compression Compact Latent Sensor Tokens Data Compression Methods Large Language Model Memory Text Analysis and Detection

Sep 1, 2026cs.AI

A Closed-Loop Evaluation of Capability Loss and Recovery in Compressed Driving Policies

Many automobile and mobility companies deploy learned driving policies on embedded computers with limited memory and power. Pruning, knowledge distillation, and quantization are the standard methods to reduce the size and the inference cost of these policies. However, these methods are commonly assessed by aggregate numerical scores, and such scores may not reflect the ability of the policy to drive safely when interacting with other road users. In this study, we propose a stage-wise closed-loop evaluation approach to follow a driving policy through a compression pipeline. We formulate the driving task as a partially observable Markov decision process (POMDP) and train a belief-state policy with proximal policy optimization (PPO) in Gym-Duckietown. We then extract the actor, compress it one stage at a time, and evaluate it on five driving curricula. We show that structured pruning is the stage at which the driving capability is first lost. Meanwhile, distillation improves the pruned actor, but the improvement is limited by its rehearsal data. Integer quantization of the improved actor loses some of the curricula that require the vehicle to stop and then resume. Interestingly, the same procedure on the unpruned actor preserves all five curricula. Our study thus provides an empirical analysis aiming to answer the currently active discussions on how to accept a compressed driving policy, so as to achieve a safe and statistically reliable deployment of automated driving functions.

Ahmad Alfan Alfian Irfan, Nur Ahmad Khatim, Mansur Arief

Information Technology, Universitas Muhammadiyah Yogyakarta, Indonesia · Division of Information Science, Nara Institute of Science and Technology (NAIST), Japan · Industrial and Systems Engineering (ISE) Department and IRC for Smart Mobility and Logistics (IRC-SML), King Fahd University of Petroleum and Minerals (KFUPM), Saudi Arabia

Proximal Policy Optimization Closed Loop Data Compression Methods

Aug 31, 2026cs.AI

Measure Before You Manage: Evaluating Agent Working Memory in Coding Agents

Agent working memory is heterogeneous. Objects such as instructions, artifacts, tool outputs, and agent-generated state play different semantic roles and exhibit different size, retention, and representation profiles. Recent work has begun to explore memory-management mechanisms that account for such heterogeneity. This work focuses on semantic heterogeneity and studies how it should shape the management and evaluation of working memory in coding agents. Across 55 archived coding-agent trajectories, we find that semantically different working-memory objects exhibit distinct retention and compression behavior. This heterogeneity motivates semantically informed memory management. We study two semantically informed strategies: an object-aware compression policy and a retrieval-based policy. Their evaluation shows that calibration gains may not transfer to held-out tasks, and that equal token budgets do not imply equal delivered context or management cost. A real-system replay further exposes serving limits that nominal budgets alone do not capture. Together, these results show why semantic structure matters for agent working memory and why evaluating memory-management strategies requires more than a nominal token budget. We organize these lessons into four levels: stored state, delivered context, management work, and task or process outcome.

Le Chen, Zishen Wan, Baixi Sun +6

Argonne National Laboratory · Columbia University · University of Houston

Working Memory Long-Term Agent Memory Coding Agents Retention Measures Data Compression Methods+1

Aug 31, 2026cs.CL

TopoCompress: Long Context Compression via Graph-Wired Semantic Trajectories

Long-context compression is essential for reducing the cost and latency of large language model inference. However, existing methods can fragment important evidence, require additional training or alignment, and often depend on the target model for effective compression. We introduce TopoCompress, a training-free and model-agnostic framework that compresses long contexts by selecting coherent semantic spans. TopoCompress first scores each span using dense and lexical query relevance together with semantic acceleration. It then constructs a hybrid graph that connects spans based on semantic similarity and sequential adjacency, and propagates the query-guided relevance scores over the graph. Across five long-context tasks-HotpotQA, 2WikiMQA, MuSiQue, Qasper, and MultiFieldQA-en-TopoCompress consistently outperforms strong compression baselines. Notably, TopoCompress achieves performance comparable to the strongest baseline while using a 4x smaller compression budget, and provides a 1.41x smaller compression time over the fastest baseline.

Daniel Agyei Asante, Yang Li

1Iowa State University, United States

Efficient Long-Context Inference LLM Inference Optimization Data Compression Methods Question-Answering Benchmarks

Aug 31, 2026cs.AI

SkillZip Pro: Execution-Aware Dynamic Compression of Progressively Loaded Skills for Self-Evolving Agents

Production agent skills are directory bundles, not isolated prompts. The root is loaded at activation; references, schemas, scripts, assets, and nested subskills are loaded only when an execution path needs them. Compressing only the root misses most deployment cost and may move branch-specific details into the always-loaded context. Flattening instead destroys progressive-loading boundaries. We introduce \method, an evaluation-free compressor for complete, progressively loaded skill bundles. It leaves the agent harness unchanged and emits an ordinary directory. The method combines two safeguards. First, it compresses \emph{across files}, removing content from a reference or subskill when the root or a declared environment contract already provides it. Second, it preserves routing, so every required file and directly callable entry remains reachable after rewriting. Users can configure \method along two independent axes. \emph{One-Shot} mode rebuilds the full bundle; \emph{Continual} mode reuses state and applies Zip-on-Write after each evolution patch. \emph{Persistent} compression rewrites the shipped bundle to reduce storage and runtime context. \emph{Transient} compression keeps that bundle byte-identical and builds a task-specific view, reducing only per-run context after build cost. Entry contracts mark private, public, and conditional resources; a multi-entry audit preserves standalone public subskills. On a production content-moderation skill evaluated by our industrial multi-round harness, \method removes \hl{38%} of skill bundle tokens and \hl{10.4%} of end-to-end per-run tokens with no quality loss, while an unprotected 71% configuration loses up to 26 accuracy points to one-sided false positives. On a multi-entry bundle, \method effeciently reduces token cost while near-perfectly preserving every route and public entry.

Xiaofan Bai, Chao Liu, Hongqiang Lin +5

1Alibaba Group · 2Zhejiang University · †Project leader

Unsupervised Skill Discovery Data Compression Methods Skills Content Moderation

Aug 31, 2026cs.LG

Tensor Methods for Language Models: From Token Representation to Training, Adaptation, Inference, Compression, and Interpretability

Large language models (LLMs) are built from structured high-dimensional objects such as token representations, weights, adaptation updates, caches, and activations, whose multilinear structure is underexploited by the conventional matrix-centric view. Tensor decompositions and tensor networks provide a principled algebraic language for this structure, yet the literature often treats them as isolated compression mechanisms. This survey organizes tensor methods for LLMs through two complementary views: a seven-stage lifecycle taxonomy covering tokenization, embeddings, pre-training, adaptation, compression, inference, and interpretability, and a component view covering embeddings, attention, and feed-forward networks. We provide unified notation and theoretical foundations, analyze tensorization strategies for individual Transformer components, and compare methods at each lifecycle stage while making differences in evaluation protocols and model scales explicit. We further connect tensor methods to neighboring efficiency techniques and probabilistic tensor networks. Finally, we synthesize open challenges and introduce

ρ_{\rm gap}

, a metric for the compression-realization gap between theoretical memory reduction and measured system-level speedup. By treating tensorization as a common structural principle, the survey provides a structured entry point to tensorized language models and clarifies when parameter savings can plausibly translate into memory efficiency, computational efficiency, or interpretability. The GitHub page dedicated to this paper is accessible at \href{https://github.com/ma-tt-a/awesome-tensor-methods-for-llms}{this https URL}.

Matvei Tarasov, Salman Ahmadi-Asl, Andre L. F. de Almeida +1

Innopolis University, Innopolis 420500, Russia · Department of Teleinformatics Engineering, Federal University of Cear´a, Fortaleza 60455-760, Brazil · Systems Research Institute of Polish Academy of Science and Warsaw University of Technology, Poland

Large Language Model Compression Tensor Networks Tensor Decomposition Data Compression Methods Interpretability Peak Memory