Visual Token Compression

Follow this topic Ask about this topic

Momentum

20 papers in the last four weeks, up 67% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 184

May 8, 2026cs.CV

An Efficient Token Compression Framework for Visual Object Tracking

Refining visual representations by eliminating their internal feature-level redundancy is crucial for simultaneously optimizing the performance and computational cost of models in visual tracking. To enhance their performance, many contemporary Transformer-based trackers leverage a larger number of historical template frames to capture richer spatio-temporal cues. However, this strategy leads to a massive number of input visual tokens. This creates two critical issues: it imposes a quadratic computational burden and can also degrade the tracker's overall performance. To bridge this gap, we propose a compress-then-interact tracking framework, ETCTrack, that learns to efficiently compress template tokens from historical template frames into a robust target representation, moving beyond handcrafted rules. Our method first employs the Adaptive Token Compressor to dynamically construct compact yet highly discriminative template tokens by filtering out redundant visual tokens. These refined template tokens are then processed by our Hierarchical Interaction Encoder to achieve a deep, adaptive interaction with the search features. Refined search features ensure subsequent precise target localization. Experiments on seven benchmarks demonstrate that our method outperforms current state-of-the-art trackers. ETCTrack-B224 reduces the number of template tokens by 60%, leading to a 21.4% reduction in MACs with only a 0.4% drop in accuracy. The source code are available at https://github.com/PJD-WJ/ETCTrack.

Weijing Wu, Qihua Liang, Bineng Zhong +3

Key Lab of Education Blockchain and Intelligent Technology, Ministry of Education, Guangxi Normal University, Guilin, 541004, China · Guangxi Key Lab of Multi-Source Information Mining and Security, Guangxi Normal University, Guilin, 541004, China · University Engineering Research Center of Educational Intelligent Technology, Guangxi Normal University, Guilin, 541004, China +1

Visual Object Tracking Efficient ViTs Visual Token Pruning Visual Token Compression

May 8, 2026cs.CV

One Token Per Frame: Reconsidering Visual Bandwidth in World Models for VLA Policy

Vision-language-action (VLA) models can use visual prediction to anticipate future states, but dense visual features make the generative sequence grow with the number of camera views, prediction horizon, and encoder resolution. Whether such dense representations are necessary for effective control remains unclear. We introduce OneWM-VLA, which represents each retained camera view with one predictive token per future step. Adaptive Attention Pooling compresses visual features into compact latents, which are jointly generated with robot actions under a conditional flow-matching objective. Future observations provide the latent targets during training and are not required at inference. This design incorporates visual prediction into a pretrained VLA policy while keeping the generative sequence compact. On MetaWorld~MT50, OneWM-VLA improves the average success rate of the

π_0

backbone from

47.91\%

to

61.53\%

, reaching

72.01\%

after 60k training steps. It also achieves

98.1\%

success on LIBERO and raises Fold Cloth success on a real Piper arm from

20.0\%

to

60.0\%

relative to

π_0

. Comparisons on two additional VLA backbones consistently favor one token over three across the evaluated checkpoints. A matched ablation at a longer action horizon further shows that removing the latent loss reduces success from

58.09\%

to

21.64\%

, supporting the benefit of future supervision for policy learning.

Zuojin Tang, Shengchao Yuan, Xiaoxin Bai +4

Zhejiang University · Central South University · Harbin Institute of Technology +2

Action-Conditioned World Models Efficient VLA Models Vision-Language-Action Models Latent World Models VLMs for Robotics Visual Token Compression

May 8, 2026cs.CV

Semantic-Aware Adaptive Visual Memory for Streaming Video Understanding

Online streaming video understanding requires models to process continuous visual inputs and respond to user queries in real time, where the unbounded stream and unpredictable query timing turn memory management into a central challenge. Existing methods typically compress visual tokens via visual similarity heuristics, or augment compression with KV-cache-level retrieval. However, compression decisions rarely incorporate semantic signals, and retrieval is often added after compression is finalized, making the two stages hard to coordinate. We present SAVEMem, a training-free dual-stage framework that brings semantic awareness into memory generation and lets the retrieval scope adapt per query. In Stage1, SAVEMem builds a three-tier streaming memory online under a constant memory budget. A fixed pseudo-question bank provides a lightweight semantic prior, so that long-term retention is shaped by semantic salience rather than visual similarity alone. In Stage2, SAVEMem performs query-aware retrieval over this memory. An anchor-conditioned recency gate adapts the retrieval scope from short-term to mid- and long-term memory based on whether the query targets the present or the distant past. Within this scope, late interaction between query and memory tokens selects candidate frames for answering. Applied to Qwen2.5-VL without training, SAVEMem improves the OVO-Bench overall score from 52.27 to 62.69 and yields consistent gains on StreamingBench and ODV-Bench, while reducing peak GPU memory by 48% at 128 frames over the backbone.

Hang Wu, Sherin Mary Mathews, Yujun Cai +2

University of California, Merced · US Bank · University of Queensland

Streaming Video Understanding Efficient VLM Inference Video QA Memory-Augmented Video Understanding Visual Token Compression

May 8, 2026cs.CV

TTF: Temporal Token Fusion for Efficient Video-Language Model

Video-language models (VLMs) face rapid inference costs as visual token counts scale with video length. For example, 32 frames at

448{\times}448

resolution already yield >8,000 visual tokens in Qwen3-VL, making LLM prefill the dominant throughput bottleneck. Existing methods often rely on global similarity or attention-guided compression, incurring offsets to their gains. We propose \textbf{Temporal Token Fusion (TTF)}, a training-free, plug-and-play pre-LLM token compression framework that exploits structured temporal redundancy in video. TTF automatically selects an anchor frame, then for each subsequent frame, performs a local window similarity search (e.g.,

3\times 3

), fusing tokens that exceed a threshold. The compressed sequence maintains positional consistency across both prefill and decoding through coordinate realignment, enabling seamless integration with existing VLM pipelines. On Qwen3-VL-8B with threshold t=0.70, TTF removes about 67% of visual tokens while retaining 99.5% of the baseline accuracy and introducing only

{\approx}0.16

,GFLOPs of matching overhead. Overall, TTF offers a practical, efficient solution for video understanding. The code is available at https://github.com/Cominder/ttf

Simin Huo, Ning LI

Shanghai Jiao Tong University

Token Merging Efficient VLM Inference Video-Language Models Video Token Compression Visual Token Compression

May 7, 2026cs.CV

LensVLM: Selective Context Expansion for Compressed Visual Representation of Text

Vision Language Models (VLMs) offer the exciting possibility of processing text as rendered images, bypassing the need for tokenizing the text into long token sequences. Since VLM image encoders map fixed-size images to a fixed number of visual tokens, varying rendering resolution provides a fine-grained compression knob. However, accuracy deteriorates quickly as compression increases: characters shrink below the vision encoder's effective resolution, making them indistinguishable. To address this, we propose LensVLM, an inference framework and post-training recipe that enables VLMs to scan compressed images, then selectively expand only the relevant images to their uncompressed form via learned tools. Building on Qwen3.5-9B-Base, LensVLM maintains accuracy comparable to the full-text upper bound at 4.3

\times

effective compression and outperforms retrieval-based, text- and visual-compression baselines up to 10.1

\times

effective compression across seven text QA benchmarks. LensVLM also generalizes to multimodal document and code understanding tasks, with the accuracy gain over baselines growing as compression increases. Our analysis validates this approach: training makes visual compression robust to rendering choices, and as compression grows the model increasingly relies on expanded content rather than unreliable visual reading. The analysis also yields practical tool-choice guidance: text expansion is preferable for rendered text, while high-resolution image expansion suits native documents whose layout cues carry task-relevant information.

Roy Xie, Dan Friedman, Donghan Yu +7

Apple · Duke University

Document Understanding Tool Use in VLMs Document Image Understanding Visual Token Compression VLM Inference

May 7, 2026cs.CV

Spark3R: Asymmetric Token Reduction Makes Fast Feed-Forward 3D Reconstruction

Feed-forward 3D reconstruction models based on Vision Transformers can directly estimate scene geometry and camera poses from a small set of input images, but scaling them to video inputs with hundreds or thousands of frames remains challenging due to the quadratic cost of global attention layers. Recent token-merging methods accelerate these models by compressing the token sequence within the global attention layers, but they apply a uniform reduction to query tokens and key-value tokens, ignoring their functionally distinct roles in 3D reconstruction. In this work, we identify a key property of feed-forward 3D reconstruction models: query tokens encode view-specific geometric requests and are sensitive to compression, while key-value tokens represent shared scene context and tolerate aggressive compression. Guided by this insight, we propose Spark3R, a training-free acceleration framework that decouples the compression of query tokens and key-value tokens by assigning distinct reduction factors, with intra-group token merging applied to query tokens and lightweight token pruning to key-value tokens. Additionally, Spark3R adaptively adjusts the key-value reduction factor across layers, further improving the quality-efficiency trade-off. As a plug-and-play framework requiring no retraining, Spark3R integrates directly into multiple pretrained feed-forward 3D reconstruction models, including VGGT,

π^3

, Depth-Anything-3, and VGGT-

Ω

, and achieves up to

28\times

speedup on 1,000-frame inputs while maintaining competitive reconstruction quality.

Zecheng Tang, Jiaye Fu, Qiankun Gao +5

School of Electronic and Computer Engineering, Peking University, Shenzhen 518055, China

3D Reconstruction Efficient ViTs Feed-Forward 3D Reconstruction Visual Token Pruning Visual Token Compression

May 7, 2026cs.CV

Learning Discrete Autoregressive Priors with Wasserstein Gradient Flow

Discrete image tokenizers are commonly trained in two stages: first for reconstruction, and then with a prior model fitted to the frozen token sequences. This decoupling leaves the tokenizer unaware of the model that will later generate its tokens. As a result, the learned tokens may preserve image information well but still be difficult for an autoregressive (AR) prior to predict from left to right. We analyze this mismatch using Tripartite Variational Consistency (TVC), which decomposes latent-variable learning into three consistency conditions: conditional-likelihood consistency, prior consistency, and posterior consistency. TVC shows that two-stage training preserves the reconstruction side but leaves prior consistency outside the tokenizer objective: the overall token distribution is fixed before the AR prior participates in training. Motivated by this view, we add a distribution-level prior-matching signal during tokenizer training, while keeping the reconstruction objective unchanged. We optimize this signal with a Wasserstein-gradient-flow update. For hard categorical tokens, the update reduces to a token-level contrast between an auxiliary AR model that tracks the tokenizer's current token distribution and the target AR prior. It requires only forward passes through the two AR models and does not backpropagate through either of them. The resulting tokenizer, wAR-Tok, reduces AR loss and improves generation FID on CIFAR-10 and ImageNet at comparable reconstruction quality.

Bowen Zheng, Yihong Luo, Tianyang Hu

The Chinese University of Hong Kong, Shenzhen · Hong Kong University of Science and Technology

Image Tokenization Visual Tokenization Autoregressive Image Generation Visual Token Compression

May 7, 2026cs.LG

VisMMOE: Exploiting Visual-Expert Affinity for Efficient Visual-Language MoE Offloading

Large-scale vision-language mixture-of-experts (VL-MoE) models provide strong multimodal capability, but efficient deployment on memory-constrained platforms remains difficult. Existing MoE offloading systems are largely designed for text-centric workloads and become much less effective for visual-heavy inputs, where large numbers of visual tokens induce broader and less predictable expert accesses. We present VisMMoE, a VL-MoE offloading system built on a single systems insight: pruning redundant visual tokens can improve offloading not only by reducing computation, but also by reshaping expert demand. We refer to this effect as \textit{visual-expert affinity}: token pruning makes expert accesses more concentrated within layers and more stable across layers, producing a smaller and more predictable expert working set. Guided by this insight, VisMMoE combines affinity-aware token compression, lookahead expert prediction, and cache/pipeline orchestration to improve expert locality and prefetch effectiveness under tight memory budgets. We implement VisMMoE on multiple frameworks and evaluate it on representative VL-MoE models and benchmarks. VisMMoE improves end-to-end inference performance by up to 2.68x and 1.61x, respectively, over strong baselines for today's VL-MoE deployments while maintaining competitive accuracy.

Cheng Xu, Xiaofeng Hou, Jiacheng Liu +1

Shanghai Jiao Tong University Minhang Qu, Shanghai, China

Expert Offloading Vision-Language Models Efficient VLM Inference Mixture-of-Experts Inference Visual Token Compression

May 7, 2026cs.CV

VideoRouter: Query-Adaptive Dual Routing for Efficient Long-Video Understanding

Video large multimodal models increasingly face a scalability bottleneck: long videos produce excessively long visual-token sequences, which sharply increase memory and latency during inference. While existing compression methods are effective in specific settings, most are either weakly query-aware or apply a fixed compression policy across frames, proving suboptimal when visual evidence is unevenly distributed over time. To address this, we present VideoRouter, a query-adaptive dual-router framework built on InternVL for budgeted evidence allocation. The Semantic Router predicts the dominant allocation policy, choosing between broad temporal coverage and adaptive high-resolution preservation, while the Image Router uses early LLM layers to score frame relevance. This enables aggressive compression on less relevant frames while preserving detail on critical evidence frames. To train both routers, we build Video-QTR-10K for allocation-policy supervision and Video-FLR-200K for frame-relevance supervision. Experiments on VideoMME, MLVU, and LongVideoBench show that VideoRouter consistently improves over the InternVL baseline under comparable or lower budgets, achieving up to a 67.9% token reduction.

Kuanwei Lin, Wenhao Zhang, Ge Li

School of Electronic and Computer Engineering, Peking University

Efficient VLM Inference Long-Video Understanding Video Token Compression Visual Token Compression Query-Aware Video Frame Selection

May 6, 2026cs.CV

Visual Text Compression as Measure Transport

Visual text compression (VTC) promises efficient long-context processing by rendering text into an image and re-encoding it with a vision-language model, often producing

3

--

20\times

fewer decoder tokens than subword tokenization. Yet token savings do not translate predictably into downstream utility: on some tasks the visual path matches or exceeds the text path, on others it collapses, and the compression ratio itself does not predict which regime will occur. The missing quantity is therefore not another summary of efficiency, but a principled measure of task-relevant information loss induced by visual encoding. We address this problem by formulating VTC in the language of measure transport. Treating text and visual tokens as empirical probability measures, we show that the ViT patch encoder induces a push-forward map whose transport cost decomposes into a precision cost from within-patch aggregation and a coverage cost from cross-patch fragmentation. Both terms are estimable from downstream-label-free probes. This formulation yields two operational consequences: a downstream-label-free routing criterion that selects whether to use the visual path for a given input or benchmark instance, and a transport-informed foveation mechanism that re-encodes high-cost regions at higher resolution. Across

24

NLP datasets at Qwen3-4B, our label-free rule matches the per-dataset oracle on

17/24

datasets (

70.8\%

), and improves the average task score by

+3.3\%

with

-10.3\%

average tokens relative to a pure-LLM.

Lv Tang, Tianyi Zheng, Yang Liu +2

University of Alberta1 · vivo Mobile Communication Co., Ltd2 · Tsinghua University3

Efficient VLM Inference Visual Tokenization Visual Token Compression

May 6, 2026cs.CV

ViTok-v2: Scaling Native Resolution Auto-Encoders to 5 Billion Parameters

Vision Transformer (ViT) autoencoders have emerged as compelling tokenizers for images, offering improved reconstruction over convolutional tokenizers. However, existing ViT tokenizers cannot explore this landscape as performance degrades outside training resolutions, and reliance on adversarial losses prevents stable scaling. ViTok (Hansen-Estruch et al., 2025) found that the compression ratio r mediates a reconstruction-generation trade-off where lower r means better reconstructions but harder generations, so improving tokenizer reconstruction is key to more Pareto-optimal tokenizers. We introduce ViTok-v2, which addresses these limitations with native resolution support via NaFlex for generalization across resolutions and aspect ratios, and a novel DINOv3 perceptual loss that replaces both LPIPS and GAN objectives for stable training at any scale. ViTok-v2 is trained on about 2B images and scaled to 5B parameters, the largest image autoencoder to date. ViTok-v2 matches or exceeds state-of-the-art reconstruction at 256p and outperforms all baselines at 512p and above. In joint scaling experiments with flow matching generators, we show that scaling both the autoencoder and the generator advances the Pareto frontier of this trade-off.

Philippe Hansen-Estruch, Jiahui Chen, Vivek Ramanujan +9

University of Texas, Austin · University of Washington · 3Stanford University +3

Image Tokenization Autoencoders Perceptual Loss Visual Tokenization Image Reconstruction Visual Token Compression

May 4, 2026cs.AI

CoVSpec: Efficient Device-Edge Co-Inference for Vision-Language Models via Speculative Decoding

Vision-language models (VLMs) have demonstrated strong capabilities in multimodal perception and reasoning. However, deploying large VLMs on mobile devices remains challenging due to their substantial computational and memory demands. A practical alternative is device-edge co-inference, where a lightweight draft VLM on the mobile device collaborates with a larger target VLM on the edge server via speculative decoding. Nevertheless, directly extending speculative decoding to VLMs suffers from severe inefficiency due to excessive visual-token computation and high communication overhead. To address these challenges, we propose CoVSpec, an efficient collaborative speculative decoding framework for VLM inference. Specifically, we first develop a training-free visual token reduction framework that prunes redundant visual tokens on the mobile device by jointly considering query relevance, token activity, and low-rank dependency. Moreover, we design an adaptive drafting strategy that dynamically adjusts both the verification frequency and the draft length. In addition, we introduce a parallel branching mechanism with decoupled verification-correction to improve draft-side utilization during target-side verification and reduce correction-related transmission overhead. Experiments on multiple benchmarks show that CoVSpec achieves up to 2.21x higher throughput than target-only inference and reduces communication overhead by more than 96% compared with baselines, without compromising task accuracy.

Yuanyuan Jia, Shunpu Tang, Qianqian Yang

College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China

Efficient VLM Inference Edge Computing Speculative Decoding Visual Token Compression

May 1, 2026cs.CV

Make Your LVLM KV Cache More Lightweight

Key-Value (KV) cache has become a de facto component of modern Large Vision-Language Models (LVLMs) for inference. While it enhances decoding efficiency in Large Language Models (LLMs), its direct adoption in LVLMs introduces substantial GPU memory overhead due to the large number of vision tokens processed during the prefill stage. To tackle this problem, we propose LightKV, a novel approach that reduces KV cache size by exploiting the redundancy among vision-token embeddings. Guided by text prompts, LightKV employs cross-modality message passing to aggregate informative messages across vision tokens and progressively compress them during prefill. This prompt-aware guidance distinguishes our method from prior vision-only compression strategies. We evaluate LightKV on eight open-source LVLMs across eight public benchmark datasets, e.g., MME and SeedBench. Experimental results demonstrate that with only 55% of the original vision tokens, LightKV (a) halves the vision-token KV cache size, (b) reduces computation by up to 40%, and (c) preserves general-purpose performance while significantly outperforming existing baselines.

Xihao Chen, Yangyang Guo, Roger Zimmermann

Integrative Sciences and Engineering Programme, National University of Singapore · School of Computing, National University of Singapore

KV Caching Efficient VLM Inference Large Vision-Language Models KV-Cache Compression Visual Token Compression

May 1, 2026cs.CV

End-to-End Autoregressive Image Generation with 1D Semantic Tokenizer

Autoregressive image modeling relies on visual tokenizers to compress images into compact latent representations. We design an end-to-end training pipeline that jointly optimizes reconstruction and generation, enabling direct supervision from generation results to the tokenizer. This contrasts with prior two-stage approaches that train tokenizers and generative models separately. We further investigate leveraging vision foundation models to improve 1D tokenizers for autoregressive modeling. Our autoregressive generative model achieves strong empirical results, including a state-of-the-art FID score of 1.48 without guidance on ImageNet 256x256 generation.

Wenda Chu, Bingliang Zhang, Jiaqi Han +4

ByteDance Seed · California Institute of Technology · Stanford University

Image Tokenization Visual Tokenization Autoregressive Image Generation Visual Token Compression

May 1, 2026cs.CV

RTPrune: Reading-Twice Inspired Token Pruning for Efficient DeepSeek-OCR Inference

DeepSeek-OCR leverages visual-text compression to reduce long-text processing costs and accelerate inference, yet visual tokens remain prone to redundant textual and structural information. Moreover, current token pruning methods for conventional vision-language models (VLMs) fail to preserve textual fidelity due to improper compression mechanisms. By analyzing the decoding process of DeepSeek-OCR, we find that a distinct two-stage reading trajectory: the model initially prioritizes the majority of high-norm tokens, then subsequently redistributes its attention to the remaining ones. Motivated by this insight, we propose RTPrune, a two-stage token pruning method tailored for DeepSeek-OCR. In the first stage, we prioritize high-norm visual tokens that capture salient textual and structural information. In the second stage, the remaining tokens are paired and merged based on optimal transport theory to achieve efficient feature aggregation. We further introduce a dynamic pruning ratio that adapts to token similarity and textual density for OCR tasks, enabling a better efficiency-accuracy trade-off. Extensive experiments demonstrate state-of-the-art performance, as evidenced by 99.47% accuracy and 1.23

\times

faster prefill on OmniDocBench, achieved with 84.25% token retention when applied to DeepSeek-OCR-Large.

Ben Wan, Yan Feng, Zihan Tang +4

Shanghai Jiao Tong University · University of Science and Technology of China · Tsinghua University

Efficient VLM Inference Optical Character Recognition Visual Token Pruning Document Image Understanding Visual Token Compression

Apr 29, 2026cs.LG

Progressive Semantic Communication for Efficient Edge-Cloud Vision-Language Models

Deploying Vision-Language Models (VLMs) on edge devices remains challenging due to their substantial computational and memory demands, which exceed the capabilities of resource-constrained embedded platforms. Conversely, fully offloading inference to the cloud is often impractical in bandwidth-limited environments, where transmitting raw visual data introduces substantial latency overhead. While recent edge-cloud collaborative architectures attempt to partition VLM workloads across devices, they typically rely on transmitting fixed-size representations, lacking adaptability to dynamic network conditions and failing to fully exploit semantic redundancy. In this paper, we propose a progressive semantic communication framework for edge-cloud VLM inference, using a Meta AutoEncoder that compresses visual tokens into adaptive, progressively refinable representations, enabling plug-and-play deployment with off-the-shelf VLMs without additional fine-tuning. This design allows flexible transmission at different information levels, providing a controllable trade-off between communication cost and semantic fidelity. We implement a full end-to-end edge-cloud system comprising an embedded NXP i.MX95 platform and a GPU server, communicating over bandwidth-constrained networks. Experimental results show that, at 1 Mbps uplink, the proposed progressive scheme significantly reduces network latency compared to full-edge and full-cloud solutions, while maintaining high semantic consistency even under high compression. The implementation code will be released upon publication at https://github.com/open-ep/ProSemComVLM.

Cyril Shih-Huan Hsu, Wig Yuan-Cheng Cheng, Chrysa Papagianni

Informatics Institute, University of Amsterdam, The Netherlands. · Open-EP Community, Taoyuan City 330049, Taiwan (R.O.C.).

Efficient VLM Inference Edge Computing Semantic Communication Edge-Cloud Computing Visual Token Compression

Apr 27, 2026cs.CV

VibeToken: Scaling 1D Image Tokenizers and Autoregressive Models for Dynamic Resolution Generations

We introduce an efficient, resolution-agnostic autoregressive (AR) image synthesis approach that generalizes to arbitrary resolutions and aspect ratios, narrowing the gap to diffusion models at scale. At its core is VibeToken, a novel resolution-agnostic 1D Transformer-based image tokenizer that encodes images into a dynamic, user-controllable sequence of 32-256 tokens, achieving a state-of-the-art efficiency and performance trade-off. Building on VibeToken, we present VibeToken-Gen, a class-conditioned AR generator with out-of-the-box support for arbitrary resolutions while requiring significantly fewer compute resources. Notably, VibeToken-Gen synthesizes 1024x1024 images using only 64 tokens and achieves 3.94 gFID; by comparison, a diffusion-based state-of-the-art alternative requires 1,024 tokens and attains 5.87 gFID. In contrast to fixed-resolution AR models such as LlamaGen -- whose inference FLOPs grow quadratically with resolution (11T FLOPs at 1024x1024) -- VibeToken-Gen maintains a constant 179G FLOPs (63.4x efficient) independent of resolution. We hope VibeToken can help unlock the wide adoption of AR visual generative models in production use cases.

Maitreya Patel, Jingtao Li, Weiming Zhuang +2

Arizona State University · SonyAI · 3Work partially done while at SonyAI. Maitreya is now at Adobe.

Image Tokenization Visual Tokenization Image Generation Autoregressive Image Generation Visual Token Compression

Apr 20, 2026cs.CV

T-REN: Learning Text-Aligned Region Tokens Improves Dense Vision-Language Alignment and Scalability

Despite recent progress, vision-language encoders struggle with two core limitations: (1) weak alignment between language and dense vision features, which hurts tasks like open-vocabulary semantic segmentation; and (2) high token counts for fine-grained visual representations, which limits scalability to long videos. This work addresses both limitations. We propose T-REN (Text-aligned Region Encoder Network), an efficient encoder that maps visual data to a compact set of text-aligned region-level representations (or region tokens). T-REN achieves this through a lightweight network added on top of a frozen vision backbone, trained to pool patch-level representations within each semantic region into region tokens and align them with region-level text annotations. With only 3.7% additional parameters compared to the vision-language backbone, this design yields substantially stronger dense cross-modal understanding while reducing the token count by orders of magnitude. Specifically, T-REN delivers +5.9 mIoU on ADE20K open-vocabulary segmentation, +18.4% recall on COCO object-level text-image retrieval, +15.6% recall on Ego4D video object localization, and +17.6% mIoU on VSPW video scene parsing, all while reducing token counts by more than 24x for images and 187x for videos compared to the patch-based vision-language backbone. The code and model are available at https://github.com/savya08/T-REN.

Savya Khosla, Sethuraman T, Aryan Chadha +2

University of Illinois Urbana-Champaign

VLM Adaptation Open-Vocabulary Semantic Segmentation Vision-Language Alignment Visual Token Compression

Apr 20, 2026cs.CV

Geometry-Guided 3D Visual Token Pruning for Video-Language Models

Multimodal large language models have demonstrated remarkable capabilities in 2D vision, motivating their extension to 3D scene understanding. Recent studies represent 3D scenes as 3D spatial videos composed of image sequences with depth and camera pose information, enabling pre-trained video-language models to perform 3D reasoning tasks. However, the large number of visual tokens in spatial videos remains a major bottleneck for efficient inference and context management. Existing pruning methods overlook the view consistency of spatial videos and the spatial diversity of the remaining tokens, which prevents them from effectively removing inter-frame redundancy and preserving scene completeness. In this paper, we propose Geo3DPruner, a Geometry-Guided 3D Visual Token Pruning framework. Geo3DPruner first models cross-frame relevance through geometry-aware global attention, and then performs a two-stage pruning process. The intra-voxel stage selects representative multi-view features within each voxel, while the inter-voxel stage preserves spatial diversity by selecting a globally distributed subset of voxels. Extensive experiments on multiple 3D scene understanding benchmarks demonstrate that Geo3DPruner retains over 90% of the original performance while pruning 90% of visual tokens, significantly outperforming existing text-guided and vision-guided pruning methods.

Han Li, Zehao Huang, Jiahui Fu +2

Beihang University · Zhongguancun Academy

Visual Token Pruning Visual Token Compression 3D Scene Understanding 3D VLMs

Apr 18, 2026cs.CV

EvoComp: Learning Visual Token Compression for Multimodal Large Language Models via Semantic-Guided Evolutionary Labeling

Recent Multimodal Large Language Models (MLLMs) have demonstrated strong performance on vision-language understanding tasks, yet their inference efficiency is often hampered by the large number of visual tokens, particularly in high-resolution or multi-image scenarios. To address this issue, we propose EvoComp, a visual token compression framework that significantly reduces token count while preserving task accuracy. EvoComp introduces a lightweight encoder-only transformer-based compressor that selects the most informative and non-redundant visual tokens by jointly considering visual and textual contexts. A core challenge lies in providing effective supervision for training the compressor. To this end, we design an evolutionary labeling strategy that searches for token subsets minimizing the MLLM's output loss, while enforcing semantic diversity through vocabulary-based token grouping. We further train the compressor using a tailored loss function combining the GHM loss to mitigate class and difficulty imbalance, and a cosine similarity regularization to encourage semantic separation between retained and discarded tokens. Extensive experiments across multiple vision-language benchmarks show that EvoComp outperforms existing methods based on attention or similarity heuristics. Notably, it retains 99.3% of the original accuracy under 3x token compression and delivers up to 1.6x speedup on mobile devices.

Jiafei Song, Fengwei Zhou, Jin Qu +7

OPPO CTG

Efficient Multimodal Inference Multimodal Large Language Models Multimodal Token Compression Visual Token Compression

Apr 17, 2026cs.AI

Why Training-Free Token Reduction Collapses: The Inherent Instability of Pairwise Scoring Signals

Training-free token reduction methods for Vision Transformers (ToMe, ToFu, PiToMe, and MCTF) employ different scoring mechanisms, yet they share a closely matched cliff-like collapse at high compression. This paper explains \emph{why}. We develop a diagnostic framework with two tools, ranking consistency

ρ_s

and off-diagonal correlation

ρ_\text{off}

, that decomposes the collapse into (1)a signal-agnostic error amplifier inherent to layer-wise reduction, predicting convex Pareto curves and

r_{\text{crit}} \propto 1/L

; and (2)shared reliance on \emph{pairwise} similarity signals whose ranking consistency degrades from

ρ_s{=}0.88

to

0.27

in deep layers. Pairwise rankings are inherently unstable (

O(N_p^2)

joint perturbations) while unary signals enjoy greater stability (

O(N_p)

perturbations, CLT). From three design principles derived from this diagnosis, we construct CATIS as a constructive validation: unary signals raise the trigger threshold, triage suppresses the gain. On ViT-Large at 63% FLOPs reduction, CATIS retains 96.9% of vanilla accuracy (81.0%) on ImageNet-1K where all baselines collapse to 43--65%.

Yang Shanglin

Visual Token Pruning Visual Token Compression

Apr 16, 2026cs.CV

StreamCacheVGGT: Streaming Visual Geometry Transformers with Robust Scoring and Hybrid Cache Compression

Reconstructing dense 3D geometry from continuous video streams requires stable inference under a constant memory budget. Existing

O(1)

frameworks primarily rely on a ``pure eviction'' paradigm, which suffers from significant information destruction due to binary token deletion and evaluation noise from localized, single-layer scoring. To address these bottlenecks, we propose StreamCacheVGGT, a training-free framework that reimagines cache management through two synergistic modules: Cross-Layer Consistency-Enhanced Scoring (CLCES) and Hybrid Cache Compression (HCC). CLCES mitigates activation noise by tracking token importance trajectories across the Transformer hierarchy, employing order-statistical analysis to identify sustained geometric salience. Leveraging these robust scores, HCC transcends simple eviction by introducing a three-tier triage strategy that merges moderately important tokens into retained anchors via nearest-neighbor assignment on the key-vector manifold. This approach preserves essential geometric context that would otherwise be lost. Extensive evaluations on five benchmarks (7-Scenes, NRGBD, ETH3D, Bonn, and KITTI) demonstrate that StreamCacheVGGT sets a new state-of-the-art, delivering superior reconstruction accuracy and long-term stability while strictly adhering to constant-cost constraints.

Xuanyi Liu, Chunan Yu, Deyi Ji +6

KOKONI 3D, Moxin Technology · Nanjing University of Science and Technology · University of Science and Technology of China +4

3D ViTs Streaming 3D Reconstruction KV Caching 3D Reconstruction Efficient ViTs Visual Token Compression+1

Apr 16, 2026cs.CV

Revisiting Token Compression for Accelerating ViT-based Sparse Multi-View 3D Object Detectors

Vision Transformer (ViT)-based sparse multi-view 3D object detectors have achieved remarkable accuracy but still suffer from high inference latency due to heavy token processing. To accelerate these models, token compression has been widely explored. However, our revisit of existing strategies, such as token pruning, merging, and patch size enlargement, reveals that they often discard informative background cues, disrupt contextual consistency, and lose fine-grained semantics, negatively affecting 3D detection. To overcome these limitations, we propose SEPatch3D, a novel framework that dynamically adjusts patch sizes while preserving critical semantic information within coarse patches. Specifically, we design Spatiotemporal-aware Patch Size Selection (SPSS) that assigns small patches to scenes containing nearby objects to preserve fine details and large patches to background-dominated scenes to reduce computation cost. To further mitigate potential detail loss, Informative Patch Selection (IPS) selects the informative patches for feature refinement, and Cross-Granularity Feature Enhancement (CGFE) injects fine-grained details into selected coarse patches, enriching semantic features. Experiments on the nuScenes and Argoverse 2 validation sets show that SEPatch3D achieves up to \textbf{57%} faster inference than the StreamPETR baseline and \textbf{20%} higher efficiency than the state-of-the-art ToC3D-faster, while preserving comparable detection accuracy. Code is available at https://github.com/Mingqj/SEPatch3D.

Mingqian Ji, Shanshan Zhang, Jian Yang

PCA Lab, School of Computer Science and Engineering, Nanjing University of Science and Technology · PCA Lab, School of Intelligence Science and Technology, Nanjing University

Multi-View 3D Object Detection Efficient ViTs 3D Object Detection Visual Token Compression

Mar 15, 2026cs.MM

DASH: Dynamic Audio-Driven Semantic Chunking for Efficient Omnimodal Token Compression

Omnimodal large language models (OmniLLMs) jointly process audio and visual streams, but the resulting long multimodal token sequences make inference prohibitively expensive. Existing compression methods typically rely on fixed window partitioning and attention-based pruning, which overlook the piecewise semantic structure of audio-visual signals and become fragile under aggressive token reduction. We propose Dynamic Audio-driven Semantic cHunking (DASH), a training-free framework that aligns token compression with semantic structure. DASH treats audio embeddings as a semantic anchor and detects boundary candidates via cosine-similarity discontinuities, inducing dynamic, variable-length segments that approximate the underlying piecewise-coherent organization of the sequence. These boundaries are projected onto video tokens as a soft temporally co-registered segmentation prior. Within each segment, token retention is determined by a tri-signal importance estimator that fuses structural boundary cues, representational distinctiveness, and attention-based salience, mitigating the sparsity bias of attention-only selection. This structure-aware allocation preserves transition-critical tokens while reducing redundant regions. Extensive experiments on AVUT, VideoMME, and WorldSense demonstrate that DASH maintains competitive or superior accuracy while achieving higher compression ratios compared to prior methods. Code is available at: https://github.com/laychou666/DASH.

Bingzhou Li, Tao Huang

Shanghai Jiao Tong University, Shanghai, China · Tongji University, Shanghai, China

Audio-Visual Understanding Efficient Multimodal Inference Multimodal Token Compression Video Token Compression Visual Token Compression

Mar 7, 2026cs.CV

StructSAM: Structure- and Spectrum-Preserving Token Merging for Segment Anything Models

Recent token merging techniques for Vision Transformers (ViTs) provide substantial speedups by reducing the number of tokens processed by self-attention, often without retraining. However, their direct application to the Segment Anything Model (SAM) family is nontrivial: SAM's image encoder mixes windowed and global attention, and its mask decoder relies on dense, prompt-conditioned features for precise boundary prediction. We systematically evaluate representative token-merging methods on SAM and Medical SAM in a strict off-the-shelf setting, and find that existing destination-selection heuristics can erode boundaries and leak prompt information as merge rates increase. We propose \textbf{StructSAM}, a resolution-preserving merge-unmerge framework tailored to SAM. StructSAM computes a lightweight token-energy score from first-order feature gradients, uses grid-based flatness screening to protect boundary and prompt regions, and merges tokens within flat areas toward low-energy destinations with explicit token recovery. We further provide a spectral graph coarsening view showing that score-guided merging yields bounded Laplacian spectral distortion compared to random or window-restricted baselines. Across eight natural and medical benchmarks, StructSAM reduces encoder FLOPs by 25-30% (up to 40%+ with prompt-aware merging) with minor drops in mIoU/Dice, consistently outperforming ToMe, PiToMe, ToMeSD, VidToMe, and ALGM at the same compute.

Duy M. H. Nguyen, Tuan A. Tran, Duong Nguyen +17

University of Stuttgart · DFKI · Max Planck Research School +10

Token Merging Image Segmentation Self-Attention Promptable Image Segmentation Efficient ViTs Attention Mechanisms+2

Mar 5, 2026cs.CV

Keeping the Evidence Chain: Semantic Evidence Allocation for Training-Free Token Pruning in Video Temporal Grounding

Video Temporal Grounding (VTG) localizes the temporal boundaries of query-relevant moments in long, untrimmed videos, making video-language-model prohibitively expensive. While recent training-free token pruning has shown success in video question answering, naively applying these objectives to VTG causes drastic degradation, as VTG crucially depends on boundary-sensitive evidence and cross-frame reasoning chains. We therefore identify two VTG-specific pruning principles: evidence retention, which keeps query-critical patches especially around event boundaries, and connectivity strength, which preserves cross-frame connectivity for long-range evidence aggregation. Building on these insights, we propose SemVID, a training-free pruning framework that constructs a compact yet coherent token subset with complementary semantic roles. SemVID first allocates per-frame budgets by balancing query relevance and inter-frame variation to avoid over-pruned segments, and then selects three types of tokens: object tokens for diverse query-critical evidence, motion tokens to capture meaningful transitions and serve as cross-frame relays, and context tokens for scene continuity. Extensive experiments show that SemVID achieves a strong accuracy-efficiency trade-off, retaining up to 95.4% mIoU with only 12.5% visual tokens and delivering up to a 5.8x prefill speedup, consistently outperforming prior methods under the same budgets. Our code is available at https://github.com/JiaqiLi404/SemVID

Jiaqi Li, Shuntian Zheng, Yixian Shen +4

University of Warwick, Coventry CV4 7AL, United Kingdom · University of Amsterdam, Amsterdam 1012 WX, Netherlands · Amazon AGI, The United States of America

Temporal Video Grounding Vision-Language Grounding Visual Token Pruning Visual Token Compression

Feb 27, 2026cs.CV

Stateful Token Reduction for Long-Video Hybrid VLMs

Token reduction accelerates long-video vision--language models (VLMs), but existing methods target Transformers, where reduction is treated as token pruning. We study token reduction in hybrid Mamba--Transformer VLMs and find that it is \emph{stateful}: Mamba layers maintain a recurrent state that accumulates information from earlier tokens, allowing discarded tokens to persist, so reduction behaves more like compression than dropping.We support this view with a representation-based probing method measuring how much information from discarded tokens is retained, and analyze layer-wise sparsity and cross-layer importance stability. Our findings show importance is sparse within layers but unstable across layers, making aggressive early pruning unreliable while hybrids remain robust to later reduction.Motivated by this, we propose a hybrid-aware token reduction framework with a low-to-high progressive schedule and a unified query-conditioned importance score for attention and Mamba layers. For Mamba, excluding the position-dependent decay from the recurrence produces a stronger selection signal. Across long-video benchmarks, our method achieves

3.8{\times}

--

4.2{\times}

prefilling speedups at a 25% token budget while maintaining near-baseline accuracy and improving with light finetuning. Hybrid models benefit from aggressive reduction, improving both efficiency and accuracy, whereas Transformers exhibit the standard trade-off. Our method also outperforms prior baselines on the same hybrid backbone and combines effectively with visual redundancy reduction methods.

Jindong Jiang, Amala Sanjay Deshmukh, Kateryna Chumachenko +7

NVIDIA

Efficient VLM Inference Video-Language Models Multimodal Token Compression Visual Token Compression

Feb 18, 2026cs.CV

AFFMAE: Scalable Vision Pre-Training for High-Resolution Microscopy Segmentation on Desktop Hardware

Self-supervised pretraining has transformed computer vision by enabling data-efficient fine-tuning, yet high-resolution pretraining typically requires server-scale infrastructure, limiting custom in-domain training for many research laboratories. Masked Autoencoders (MAE) reduce computation by encoding only visible tokens, but combining MAE with hierarchical downsampling architectures has remained structurally challenging due to dense grid priors and mask-aware design compromises. We introduce AFFMAE, a masking-friendly hierarchical pretraining framework built on adaptive, off-grid token merging. AFFMAE removes dense-grid assumptions while preserving hierarchical scalability during pre-training and fine-tuning. To support this architecture, we developed numerically stable mixed-precision Triton kernels and a lightweight, point-based decoder that can be directly repurposed as a segmentation head. On high-resolution microscopy segmentation, AFFMAE matches MAE finetuning performance on foot process width estimation with ViT backbone at equal parameter counts while being 2x faster during pre-training and halving peak memory usage. Furthermore, AFFMAE achieves up to 5x throughput speedups fine-tuning at the 1024px resolution, providing high-resolution model training on desktop hardware. Code available at https://github.com/najafian-lab/affmae.

David Smerkous, Zian Wang, Behzad Najafian

Department of Laboratory Medicine & Pathology, University of Washington, Seattle WA, USA · Electrical Engineering & Computer Science, Oregon State University, Corvallis, OR, USA · Paul G. Allen School of Computer Science & Engineering, University of Washington, Seattle WA, USA

Masked Autoencoders Self-Supervised Pre-Training Efficient ViTs Microscopy Image Analysis Visual Token Compression

Feb 17, 2026cs.CV

LGQ: Learnable Geometric Quantization for Image Tokenization

Recent collapse-free quantizers such as FSQ achieve stable training by replacing the learnable codebook with an engineered geometry: a fixed scalar grid whose structure is dictated by the codebook size K. We show this trade-off is unnecessary. We introduce Learnable Geometric Quantization (LGQ), which retains a learnable codebook of codes and performs soft-to-hard assignment via temperature annealing, regularized by two cheap terms: A diversity term scaled by codebook size that penalizes concentrated batch-average usage is the primary driver of collapse resistance, complemented by a peakedness term that sharpens each token's soft-assignment toward one-hot; together they prevent codebook collapse without EMA, reset heuristics, or codebook reparameterization. Under a fixed VQ-GAN backbone, we benchmark LGQ against RotVQ, FSQ, SimVQ, and IBQ on ImageNet 256x256 at K=16,384 and sweep LGQ over K={4096,..,65,536} without any per-K hyperparameter tuning. LGQ attains the best reconstruction FID at K = 16,384 while maintaining 100% codebook utilization, and continues to improve as the codebook grows to K = 65,536 with utilization unchanged. Training MaskGIT on the frozen tokenizers, LGQ further attains the best class-conditional generation among the compared quantizers, leading on reconstruction and generation alike. Code is available at https://github.com/KurbanIntelligenceLab/LGQ.

Idil Bilge Altun, Mert Onur Cakiroglu, Elham Buxton +2

Luddy School of Informatics, Computing and Engineering, Indiana University Bloomington · Department of Computer Science, University of Illinois Springfield · College of Science and Engineering, Hamad Bin Khalifa University, Doha, Qatar

Image Tokenization Vector Quantization Visual Token Compression

Jan 29, 2026cs.CL

CausalEmbed: Auto-Regressive Multi-Vector Generation in Latent Space for Visual Document Embedding

Although Multimodal Large Language Models (MLLMs) have shown remarkable potential in Visual Document Retrieval (VDR) through generating high-quality multi-vector embeddings, the substantial storage overhead caused by representing a page with thousands of visual tokens limits their practicality in real-world applications. To address this challenge, we propose an auto-regressive generation approach, CausalEmbed, for constructing multi-vector embeddings. By incorporating iterative margin loss during contrastive training, CausalEmbed encourages the embedding models to learn compact and well-structured representations. Our method enables efficient VDR tasks using only dozens of visual tokens, achieving a 30-155x reduction in token count while maintaining highly competitive performance across various backbones and benchmarks. Theoretical analysis and empirical results demonstrate the unique advantages of auto-regressive embedding generation in terms of training efficiency and scalability at test time. As a result, CausalEmbed introduces a flexible test-time scaling strategy for multi-vector VDR representations and sheds light on the generative paradigm within multimodal document retrieval. Our code is available at https://github.com/Z1zs/Causal-Embed.

Jiahao Huo, Yu Huang, Yibo Yan +7

The Hong Kong University of Science and Technology (Guangzhou) · Alibaba Cloud Computing · The Hong Kong University of Science and Technology +1

Multimodal Embedding Multi-Vector Retrieval Visual Document Retrieval Visual Token Compression