Efficient VLM Inference

VLM: Vision-Language Model

Follow this topic Ask about this topic

Momentum

38 papers in the last four weeks, up 100% on the four weeks before. 0.4% of all new papers.

Jul 13Week of Sep 28

Latest papers 322

Oct 8, 2026cs.CV

DVD: Dynamic Vector Decoding for Efficient MLLM-based Perception

Multimodal large language models have made remarkable progress in bridging vision and language, facilitating various perception tasks essential for human-machine interaction, robotics, and autonomous driving. However, existing MLLM-based perception methods predominantly rely on text-based coordinate representation, which suffers from excessive token overhead, or fixed-range quantization, which suffers from range and precision constraints, especially for 3D domains with unbounded spatial range and high localization accuracy requirements. To address these challenges, we propose a dynamic vector decoding method named DVD, which unifies the representation of 2D and 3D perception tasks. Specifically, we first transform diverse perceptual representation (i.e., 2D bounding boxes, 2D masks, and 3D bounding boxes) into 1D vector sequences, which are then mapped to compact discrete tokens in the high-dimensional space. Then, a lightweight de-tokenizer enables seamless integration with MLLMs by decoding output tokens back to original 2D and 3D perceptual representations. Extensive experiments on 2D and 3D perception benchmarks including RefCOCO series, SUN-RGBD, KITTI, Hypersim, nuScenes demonstrate that DVD achieves superior performance in 2D and 3D tasks and reduces significantly the token overhead and inference latency. DVD provides an efficient and general framework for integrating perception capabilities into MLLMs, overcoming the inherent limitations of existing methods.

Jinghua Hou, Zhe Liu, Hengshuang Zhao

The University of Hong Kong

Efficient VLM Inference Multimodal Large Language Models 3D VLMs

Oct 8, 2026cs.CV

V-CoLA: Vision Token Compression with Linear Attention

Vision-language models (VLMs) have demonstrated impressive capabilities but suffer from substantial computational overhead, as vision tokens dominate the input sequence. This motivates vision token compression as a key direction to alleviate the burden. However, with the emergence of hybrid architectures incorporating linear attention (\eg, Qwen3.5), prior methods designed for softmax attention struggle to generalize. Our analysis reveals that both attention- and similarity-based approaches suffer notable performance degradation, underscoring the urgent need for compression methods tailored to this regime. To this end, we propose \textbf{V-CoLA}, an efficient training-free token compression framework specifically designed for linear attention. V-CoLA introduces a novel \textit{uniqueness-aware importance criterion} for identifying critical vision tokens, coupled with an \textit{adaptive token merging strategy} that performs compression. All components are optimized at the implementation level to remain compatible with the chunk-wise parallelism of linear attention, ensuring strong practical value. Extensive experiments across multiple benchmarks demonstrate the superiority of V-CoLA: it achieves 99.5% of the original performance with only 50.0% of vision tokens, and over 88.0% with as few as 12.5%, while delivering a 1.86

\times

to 6.15

\times

prefill speedup.

Hao Jiang, Yiru Mao, Tianpeng Bu +9

Alibaba Cloud Computing, Alibaba Group

Visual Token Compression Efficient VLM Inference

Oct 7, 2026cs.CV

Adaptive Visual Token Reduction for Accelerated Image Understanding

Large Vision-Language Models achieve strong VQA performance, but processing high-resolution, information-rich images requires substantial computation, motivating visual token reduction. However, existing methods often prune individual tokens or rely on fixed-size cropping, limiting their ability to preserve spatially structured information such as horizontally or vertically elongated text. To address this limitation, we propose ReFIT, an instruction-guided visual token reduction framework for efficient LVLM inference. ReFIT consists of Relevance-Guided Window Reshaping (RWR) and Instruction-Guided Token Refinement (ITR), where RWR captures instruction-relevant regions by adapting to their spatial characteristics, while ITR further removes unnecessary visual tokens. Experiments on four VQA benchmarks demonstrate that ReFIT improves answer accuracy while reducing computational cost, and qualitative results demonstrate its effectiveness in localizing relevant regions and removing unnecessary visual information.

Seyoung Jeong, Jong Pil Yun, Sang Jun Lee

Jeonbuk National University, Jeonju, Republic of Korea · Korea Institute of Industrial Technology (KITECH), Incheon, Republic of Korea · Chung-Ang University, Seoul, Republic of Korea

Efficient VLM Inference Visual Token Pruning Visual Token Compression Visual Question Answering

Oct 6, 2026cs.CV

Foveated Compression: Selective High-Resolution Preservation for Token-Efficient VLMs

Visual tokens are a major source of inference cost in vision-language models, yet simple image downsampling remains a surprisingly strong compression baseline. This raises a complementary question: under a fixed token budget, where should visual fidelity be preserved? We introduce Foveated Compression, which encodes a full-resolution image once and represents it with a mixture of native- and compressed-resolution visual tokens. A behaviorally self-distilled Foveated Merger compresses local visual tokens while preserving compatibility with their native counterparts, and a lightweight Foveated Selector chooses one of nine spatial cells to retain at native resolution using exhaustive budget-matched intervention supervision. At 11.11% visual tokens, uniform Foveated Compression shows no significant paired difference from iso-token downsampling. At 20.99%, the learned selector significantly outperforms random and fixed allocation, but remains below strong whole-image resizing, showing that localized fidelity is not universally preferable. A budget-matched region-choice oracle reaches 82.73 macro accuracy versus 69.61 for the learned selector, revealing substantial headroom within the same spatial action space. Matched probing further shows that signals predicting when compression breaks the answer are substantially more accessible after language-model computation than to the lightweight prefill-free selector. These results expose complementary bottlenecks in region selection and compressed-region fidelity.

Donghyun Han, Jangho Park, Yuseok Bae

ETRI, South Korea · Kyung Hee University, South Korea

Efficient VLM Inference Visual Token Pruning Visual Token Compression

Oct 5, 2026cs.CV

Efficient Test-time Adaptation through Candidate Verification and Divergence Shifts

Vision-language models (VLMs) achieve strong zero-shot transferability but remain vulnerable to target-domain shifts at inference time. Test-time adaptation (TTA) offers a practical remedy, yet most existing VLM-TTA methods follow a prediction-side adaptation paradigm. They use test samples to adjust logits, prototypes, caches, priors, or feature statistics, often incurring additional computational overhead. In this paper, we take a different perspective and reframe VLM-TTA as candidate verification rather than prediction adjustment. We propose Test-Time Correction (TTC), a hypothesis-based correction framework guided by a simple principle: hypothesize, reconstruct, correct. Given a test feature and its top-k candidate labels, TTC treats each candidate label as a hypothesis, reconstructs the feature within the corresponding latent subspace stored in a memory bank, and measures the resulting divergence shift. This shift quantifies how much the candidate subspace and its relations to other candidates change after the hypothetical insertion of the test feature. A correct candidate hypothesis induces only a small shift, whereas an incorrect one perturbs the subspace more strongly. TTC therefore corrects the prediction by selecting the candidate with the minimum aggregated divergence shift. This training-free candidate-verification mechanism avoids iterative optimization and provides a favorable accuracy-efficiency trade-off. Across five TTA settings and 15 benchmark datasets, including zero-shot classification, domain generalization, few-shot classification, base-to-novel generalization, and cross-dataset evaluation, TTC consistently improves accuracy over state-of-the-art VLM-TTA methods while achieving up to 2x speedup, over 3x lower CPU memory usage, and up to 1.4x lower GPU memory usage than the lowest-memory training-free baseline.

Seungmin Oh, Seunghun Kang, Jongbin Ryu

Ajou University

Test-Time Adaptation Test-Time Adaptation of VLMs Efficient VLM Inference

Oct 4, 2026cs.AI

CASE: Cost-Aware Stopping for Efficient Long-Video Agents

Long-video agents can actively gather question-relevant evidence, but they typically leave a central decision implicit: when has the agent seen enough to answer? We propose CASE, a plug-in termination framework that frames this decision as policy-conditioned sequential stopping. At each causal checkpoint, CASE combines an auxiliary multiple-choice assessment of accumulated evidence with the host agent's execution state. From complete native trajectories, we construct a cost-aware target that compares answering now with stopping later along the same search path, accounting jointly for answer correctness and the full cost of continued reasoning. A lightweight Ridge regressor learns this decision gap and produces STOP/CONTINUE decisions. We evaluate three vision-language models with VideoSeek and AVP. On Video-MME, end-to-end accuracy changes by +0.67 percentage points on average while CASE reduces model-token use by 53.63%. The same frozen policies then transfer zero-shot to LongVideoBench and MLVU, with end-to-end accuracy changes of +3.38 and +4.58 points while saving 58.78% and 51.28% of model tokens, respectively. Across all agent-model-benchmark combinations, CASE attains the highest accuracy-efficiency Pareto-frontier coverage among the compared stopping methods (83.3%) at the selected operating points. Online execution preserves this favorable accuracy-efficiency trade-off and additionally reduces measured runtime by 54.1% on average. CASE provides a plug-in termination framework for long-video reasoning agents, enabling them to decide when further evidence acquisition is no longer worthwhile.

Yiming Du, Chenghao Liu, Zhiyuan Liu +4

Peking University

Efficient VLM Inference Video Reasoning Long-Video QA

Oct 1, 2026cs.CV

VETO: Video Efficient Token Optimization for Vision Language Models

Processing long videos with Vision-Language Models (VLMs) is bottlenecked by the quadratic cost of visual tokens, making long-form inference prohibitively expensive. While single-axis compression methods mitigate this, they hit a hard efficiency floor because they treat spatial and temporal redundancy independently. We present VETO (Video Efficient Token Optimization for Vision-Language Models), a training-optional plug-in that eliminates this bottleneck through dual-axis compression: (i) an intra-frame compressor that merges semantically similar tokens within each frame via optimal-transport inspired matching, and (ii) an inter-frame compressor that identifies and merges temporally redundant frames. The key design insight is hierarchical ordering: by first compressing spatial dimensions, VETO drastically reduces the cost of subsequent global temporal matching, bypassing the efficiency wall of single-axis approaches, with an advantage that grows with modern fully-fused attention infrastructure. Empirically, VETO achieves up to 45% faster inference (e.g., on LLaVA-OneVision-7B) while preserving or improving accuracy. Under extreme token starvation (10% budget), VETO outperforms VFlowOpt (54.9%), VisionZip (52.6%), and FastV (47.9%) with 55.7% accuracy. We demonstrate universal applicability across LLaVA-OneVision, InternVL-2.5, and LongVA, with zero-shot accuracy preserved or improved in all cases.

Gueter Josmy Faure, Hao Ping Wang, Min-Hung Chen +1

National Taiwan University · NVIDIA

Efficient VLM Inference Large Vision-Language Models Video Token Compression Visual Token Compression

Oct 1, 2026cs.CV

Not All Error Yields to Scale: Where Scaling Stops in Vision-Language Inference

Vision-language models (VLMs) face a fixed-budget trade-off between processing more visual information for fine-grained perception and using a larger language backbone for complex reasoning. Existing studies do not tell us which combination of backbone size and input resolution to deploy, especially in high-resolution deployments. To address this gap, we propose the Separable Law that describes how VLM performance changes with language backbone size and visual token count. We fit the law to measurements from 26 InternVL and QwenVL models, with language backbone sizes from 1B to 72B, on four high-resolution benchmarks with image sizes from 224 pixels to 8K. We find that the questions responding to scaling can be predicted from the skill they require, while a substantial fraction never responds at all. We also find that the two model families gain similarly from a larger backbone, while their gains from more visual tokens differ sharply. Combined with a cost law, the Separable Law gives a closed-form rule for allocating compute between backbone size and visual tokens. When deployment is limited to available configurations, the law identifies model and image sizes that perform close to the best feasible choice under the same budget. We hope our work offers a principled way to decide how much a model should be allowed to see at high resolution, given what it must reason about.

Xinye Zhao, Yunkai Dang, Yunchen Wu +1

School of Intelligence Science and Technology, Nanjing University

Vision-Language Models Efficient VLM Inference Language Model Scaling Laws

Oct 1, 2026cs.CV

MWOP: Modality-aware Width-wise Operation Pruning for Efficient MLLMs

Multimodal large language models (MLLMs) incur substantial inference costs when processing long visual-textual sequences. While existing operation compression methods exploit modality-level redundancy, they largely treat computation within attention heads and shared feed-forward network (FFN) channels as unified units, leaving finer-grained redundancy underexplored. We find that redundancy varies both across modality-interaction paths within the same attention head and across visual and textual executions of the same FFN channel. Based on these findings, we propose Modality-aware Width-wise Operation Pruning (MWOP), which independently prunes visual-to-visual (V2V), text-to-visual (T2V), and text-to-text (T2T) attention paths within each layer, and separately selects FFN channels for visual and textual inputs. A first-order Taylor criterion guides the pruning process, with FFN importance re-evaluated after attention pruning and LoRA-based recovery training. To translate the resulting fine-grained sparsity into practical acceleration, we further develop path-sparse Triton attention kernels and compact visual-side FFN execution. MWOP preserves the token sequence while reducing attention and FFN computation, making it complementary to token compression and enabling simultaneous reduction of sequence length and per-token computation. On LLaVA-OneVision-7B, MWOP alone achieves a

1.6\times

prefill speedup with 99.7% average performance retention across 12 benchmarks. Combined with two representative token compression methods, it further increases their prefill speedups from

2.0\times

and

1.9\times

to

2.9\times

and

2.7\times

, respectively. Results on Qwen2.5-VL-7B further demonstrate its applicability across architectures. The code is available at https://github.com/EIT-NLP/MWOP.

Xudong Wang, Hao Wu, Haozhe Hu +5

EIT-NLP Lab, Eastern Institute of Technology, Ningbo · Shanghai Jiao Tong University · The Hong Kong Polytechnic University +1

LLM Pruning Efficient VLM Inference Multimodal Large Language Models Neural Network Pruning

Sep 30, 2026cs.CV

CoVisco: Codec-Native Vision Encoder with Native Token Compression for Unified Image-Video Understanding

Vision-language models face a fundamental scaling bottleneck: the number of visual tokens grows with both temporal duration and spatial resolution, making long-video understanding expensive for the vision encoder and the language model. Existing methods often compress visual tokens after dense encoding, creating a mismatch between the representation used during training and the compact interface required at deployment. We present CoVisco, a codec-native vision encoder with native token compression for unified image-video understanding. By combining codec-native input support with segmented attention, CoVisco can encode long visual inputs in a single forward pass without forming dense patch-to-patch interactions across all frames. Each temporal segment is equipped with learnable abstract tokens that learn a compact segment-level representation, while fine-grained patch tokens remain available throughout the encoder. Alternating intra-segment and abstract-communication layers preserve video-level context through the abstract-token channel. A lightweight selector further exposes either abstract tokens alone or abstract tokens augmented with a runtime-selected subset of patch tokens, yielding a compact visual interface that reduces the visual context and prefill burden of downstream MLLMs while retaining fine-grained evidence when needed. Pretrained with contrastive objectives on 565M image--text pairs and 6.4M videos, CoVisco shows competitive performance on video-oriented embedding and multimodal understanding benchmarks. In the evaluated four-segment, 64-frame setting, abstract-only inference uses only 400 visual tokens while achieving video-understanding performance close to, and on some benchmarks exceeding, OneVision-Encoder. Selected patch tokens further improve fine-grained video reasoning. Project URL: https://github.com/ernie-research/CoVisco.git

Yulong Liu, Xiaotian Han, Junyuan Shang +6

ERNIE Team, Baidu Inc. · The Hong Kong University of Science and Technology · Institute of Automation, Chinese Academy of Sciences (CASIA)

Video Understanding Efficient VLM Inference Video Representation Learning Video Token Compression Visual Token Compression

Sep 30, 2026cs.CV

When Masking Helps or Hurts Robustness in Compressed CLIP: A Pre-Deployment Diagnostic

This paper demonstrate that whether masking-based token pruning helps or hurts worst-group robustness can be predicted before deployment, without labels or fine-tuning. A systematic study of semantic masking across 8 spurious-correlation benchmarks shows its effect on worst-group accuracy is highly unstable: it improves accuracy by up to 82.5% relative on some datasets and degrades it by up to 100% on others. We trace this instability to spurious inversion: background patches receive higher CLIP text-similarity than the true object when the spurious attribute is background-separable, inverting the assumption every text- and attention-guided pruning method relies on. We introduce the Spurious Inversion Metric (SIM), a label-free, pre-deployment diagnostic whose sign predicts this effect with statistical significance (binomial

p=0.035

) across all 8 datasets, and remains dependable across 6 CLIP architectures with a clean foreground/background split. Naive masking is itself a major source of risk: it causes the largest average-accuracy loss of any method we evaluate, and its own per-image segmentation step is a significant runtime bottleneck. To address this, we design a batched, synchronization-free GPU segmentation routine that cuts this overhead from 3.5

\times

to 1.75

\times

baseline. Gating deployment by SIM's sign recovers masking's benefits while avoiding its worst failures, matching or exceeding a strong pruning baseline on 7 of 8 datasets.

Muhammad Zawish, Steven Davy

Technological University Dublin, Ireland

VLM Robustness Efficient VLM Inference Spurious Correlation Robustness Visual Token Pruning

Sep 30, 2026cs.CV

GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed

Autoregressive (AR) grounding models serialize spatial predictions, introducing sequential latency and imposing a causal order on output tokens. We view grounding as visual evidence extraction: objects, locations, and spatial relations are jointly constrained by the image and query, yet their dependencies do not imply an intrinsic left-to-right generation order. This distinction makes bidirectional diffusion a natural fit, allowing spatial hypotheses to emerge in parallel and be jointly refined through iterative denoising. We introduce GroundAnything, a 4B-parameter grounding foundation model that reconciles fast parallel decoding with precise localization through blockwise denoising. Training combines grounding pretraining from public datasets and dedicated data engines, direct AR-to-diffusion conversion with joint AR and diffusion objectives, supervised fine-tuning, and GRPO-based reinforcement post-training. Across 30 grounding benchmarks, our autoregressive variant, GroundAnything-VLM, establishes a new overall state of the art among similarly sized models at 72.42%, remaining competitive with GPT-6 Astra (71.35%). With entropy-guided decoding, GroundAnything also surpasses the prior state of the art at this scale, averaging 61.75% versus 53.32% for the fast MTP-based LocateAnything model. We further explore decoding strategies, showing that an optional self-speculative mode achieves a

4.51\times

speedup over the AR counterpart with a 0.74 percentage-point drop in COCO F1mIoU. Infrastructure experiments show that progressive inference optimizations translate parallel decoding into practical speedups. These support efficient visual grounding in latency-sensitive real-world systems.

Qize Yu, Lianrui Fan, Bowen Ping +19

XPeng Inc. · Peking University · The University of Hong Kong +4

Efficient VLM Inference Vision-Language Grounding Parallel Decoding Diffusion Language Models

Sep 29, 2026cs.CV

TReVS: Integrating Textual Relevance and Visual Saliency for Efficient Vision-Language Model Token Pruning

Vision-Language Models (VLMs) excel at visual understanding and reasoning but often incur substantial inference costs due to the large number of visual tokens. Recent visual token pruning methods increasingly follow a two-stage paradigm: they first remove visually redundant tokens after the vision encoder and then discard tokens irrelevant to the textual query within the Large Language Model (LLM). However, since the first stage typically relies solely on vision-encoder saliency, it may prematurely eliminate query-relevant tokens, depriving the subsequent text-guided stage of critical visual evidence. Our empirical analysis shows that incorporating query guidance into first-stage pruning better preserves task-relevant evidence and consistently improves performance over vision-only saliency-based pruning. We further find that high-variance attention heads are more sensitive to the textual query and yield more discriminative text-to-vision attention signals for second-stage pruning. Motivated by these findings, we propose TReVS, a training-free framework that combines textual relevance with vision-encoder saliency for pre-LLM pruning and leverages high-variance attention heads to remove task-irrelevant tokens at shallow-to-intermediate layers of the LLM. On LLaVA-1.5-7B, TReVS retains 92.8% of the unpruned baseline performance while pruning 94.4% of visual tokens, outperforming prior state-of-the-art methods.

Jing Wang, Zhiping Wu, Dongdong Ren +3

School of Intelligence Science and Technology, Nanjing University · School of Electronic Science and Engineering, Nanjing University · Geely Automobile Research Institute (Ningbo) Co., Ltd., 315000 +1

Efficient VLM Inference Large Vision-Language Models Visual Token Pruning

Sep 29, 2026cs.CV

LazySloth: Bounded LLM-based Lazy Tree Search for Fast Long Video Comprehension

Modern vision-language models (VLMs) have shown promising results in long-video understanding due to the rich semantic information they can capture. However, most methods focus on coarse captioning of extracted image frames that are computationally inefficient and require models with large context windows. While past work has explored efficient methods through multimodal retrieval-augmented generation (RAG), they rely on lossy embeddings that lose temporal context and fine-grained detail. Few works to date have investigated how VLM-based query-relevant information retrieval can be optimized. We introduce LazySloth, an efficient tree-based search method that speeds up video comprehension and retrieval tasks 2.9-8.3x (compared to existing agentic methods) through bounded captioning of portions of the video considered irrelevant by a VLM of the video. Compared to contemporary specialized video-understanding VLMs and RAG-based methods, LazySloth achieved similar or better final task accuracy across two recent open-source base VLMs--Gemma 4 31B and Qwen3.6 27B--across four benchmarks. LazySloth reduced the gap between the base open-source model and a closed-source model, GPT-4o. Ablations showed that replacing VLM scene understanding with CLIP-based retrieval cost 8.8-19.9% in accuracy, while lazy tree construction matches eager construction at a fraction of the captioning cost. With LazySloth, we demonstrate the possibility of faster long-video comprehension without substantial loss in performance.

Arka Mukherjee, Kaleen Shrestha, Larissa Zhu +1

School of Computer Engineering, Kalinga Institute of Industrial Technology (KIIT) Bhubaneswar · Computer Science Department, Viterbi School of Engineering, University of Southern California

Efficient VLM Inference Long-Video Understanding Vision-Language Retrieval

Sep 29, 2026cs.CV

Beyond Attention Imbalance: Mitigating Hallucinations via Spectral Surgery

While Large Vision-Language Models (LVLMs) achieve remarkable success, hallucinations remain a significant barrier to their reliable deployment. Recent studies primarily attribute these issues to cross-modal attention imbalances; most solutions therefore focus on reweighting visual tokens or suppressing language priors. However, such approaches often overlook the spectral characteristics of the visual information flow and frequently rely on Contrastive Decoding (CD), which doubles inference time. Instead of following conventional approaches, we identify two distinct hallucination patterns-Perceptual-Semantic Dissociation and Localized Fixation-and propose FLASH (Frequency-Localized Attention SHaping), a training-free and CD-free framework. FLASH utilizes a Spectral Vortex Score to detect vision heads within multi-head attention layers and applies adaptive spectral modulation to rectify the visual information flow during decoding. Empirical results demonstrate that FLASH achieves a superior balance between performance and efficiency compared to SOTA methods.

Siqi Lu, Suo Wei, Yongbin Zheng +3

College of Intelligence Science and Technology, National University of Defense Technology, China · School of Computer Science, Northwestern Polytechnical University, China · Ningbo Institute, Northwestern Polytechnical University, China +1

Vision-Language Models Efficient VLM Inference Large Vision-Language Models VLM Hallucination Mitigation

Sep 29, 2026cs.CV

GleanVID: Complementary Token Selection for Efficient Video Large Language Models

Video Large Language Models (VideoLLMs) have achieved strong video understanding capabilities but incur substantial inference overhead due to the large number of visual tokens. Existing VideoLLM token compression methods largely rely on selection-independent scoring, overlooking cross-frame complementarity and consequently retaining redundant evidence across frames. Instead, we view video token selection as a progressive evidence accumulation process. It aims to retain visual evidence that is individually informative and collectively complementary under a limited token budget. Building on this insight, we introduce GleanVID, a training-free inference acceleration framework for VideoLLMs. Specifically, GleanVID first allocates the global token budget across frames according to temporal novelty and then selects tokens by jointly considering local representativeness and subspace complementarity, thereby preserving richer and less redundant visual evidence. Extensive experiments across diverse VideoLLMs and benchmarks demonstrate that GleanVID consistently achieves state-of-the-art performance. Notably, with only 25% of visual tokens, GleanVID preserves 98.6% of Qwen3-VL's original performance while reducing its prefill latency by 44.7%. On LLaVA-OV-7B, GleanVID at a 25% retention ratio even slightly surpasses the original model.

Shuo Yang, Changbai Li, Rui Tang +3

Beihang University · Communication University of China

Efficient VLM Inference Video-Language Models Video Token Compression

Sep 29, 2026cs.CV

Representation Dynamics Reveal Semantic Saliency and Similarity for Visual Token Pruning in MLLMs

Multimodal large language models (MLLMs) incur high inference latency from long visual token sequences. Existing pruning methods commonly use attention maps or output features to estimate token importance or redundancy. Several recent approaches also exploit representation changes, but when and how these changes reflect foreground saliency and semantic consistency remain insufficiently understood. We analyze visual token representation dynamics across encoder depth and uncover two findings. First, the relationship between token update magnitudes and foreground saliency is layer-dependent: large token updates concentrate on foreground regions in two depth intervals, separated by several sink-dominated layers at intermediate depths. Second, similarities between token update directions better distinguish same-class from different-class tokens than those between encoder output features. Building on these findings, we propose MSDG-Prune, a training-free method that uses update magnitudes and directions to preserve salient and diverse visual information. Specifically, we group tokens by update-direction similarity and use query-weighted saliency derived from update magnitudes across a chosen depth window for group-wise token pruning. Extensive experiments across four MLLMs demonstrate the effectiveness and generalizability of MSDG-Prune. On LLaVA-NeXT, it retains 91.9% of uncompressed performance on average with only 5.6% of visual tokens, while achieving a 7.8x prefilling speedup. Code is available at https://github.com/liweixuan-hitsz/MSDG-Prune.

Weixuan Li, Zikun Zhou, Xinyi Zhuang +4

Harbin Institute of Technology, Shenzhen · Shenzhen Loop Area Institute

Efficient VLM Inference Visual Token Pruning Visual Token Compression

Sep 29, 2026cs.CV

FocusVTC: Efficient and High-Performance Visual Text Compression with Adaptive Resolution

Long-context reasoning in large language models incurs substantial computation and memory costs. Visual text compression (VTC) reduces input length by rendering text as images, but fixed-resolution rendering creates a compression-performance trade-off: low DPI saves tokens at the expense of legibility, whereas high DPI spends tokens on irrelevant content. We introduce FocusVTC, which breaks this trade-off through adaptive resolution while preserving general multimodal capabilities. It combines compressed low-DPI global views with selective region enhancement, integrating enhanced views into ongoing reasoning. We construct 29.4K high-quality Reasoning-Evidence Localization (REL) chain-of-thought examples (REL-CoT) that link reasoning traces to page indices and bounding boxes. Multi-resolution REL supervised fine-tuning (REL-SFT) teaches the model to localize relevant regions, and Group Relative Policy Optimization learns when to enhance resolution and how to use the resulting observations, without a separate continual-pretraining stage. At 72 DPI on RULER v1, FocusVTC scores 87.4 at

2.9\times

input compression, including tool observations, versus 57.5 for Glyph at

3.0\times

input compression. It surpasses its text-input backbone on LongBench (56.40 versus 55.86), improves the MRCR macro-average by 13.91 points, and achieves a 51.19 macro-average on VTCBench. The MRCR latency evaluation also shows a

2.79\times

online end-to-end speedup over Text. General multimodal capabilities are preserved, with MMMU increasing from 65.12 to 66.73 and MME from 2424.02 to 2457.62.

FangZhi Zhong, Xuerui Qiu, Yuqi Pan +4

Institute of Automation, Chinese Academy of Sciences · School of Artificial Intelligence, University of Chinese Academy of Sciences · Zhongguancun Academy +1

Efficient VLM Inference VLM Reasoning Long-Context VLMs Visual Token Compression

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

Efficient VLM Inference Large Vision-Language Models Visual Token Compression

Sep 28, 2026cs.RO

NavJev: Efficient Vision-Language Navigation via Action-Centric Visual Compression and Discriminative Action-Semantic Memory

Recent zero-shot Vision-and-Language Navigation (VLN) methods increasingly rely on multimodal large language models (MLLMs) to reason over visual observations, navigation instructions, and candidate actions. Although effective, repeatedly invoking autoregressive multimodal reasoning at every navigation step introduces substantial inference latency, limiting the responsiveness of embodied agents. We propose NavJev, an efficient VLN framework that reformulates online navigation from repeated multimodal generation into compact visual compression followed by lightweight typed action selection. Specifically, Action-Centric Visual Compression (ACVC) integrates waypoint geometry, BLIP captions, and RAM semantic tags into compact representations of candidate actions, while Discriminative Action-Semantic Memory (DASM) filters shared semantics and maintains discriminative action-specific evidence across navigation steps. Based on these representations, Jev directly performs structured probabilistic decisions over the available action set. Experiments on R2R-CE show that NavJev achieves 27.0% SR and 22.4% SPL with only 0.65 s per navigation step, while substantially reducing inference latency and cost compared with MLLM-based VLN methods. The project page is available at https://kai-sheng-caesar.github.io/NavJev/.

Kai Sheng, Liuyi Wang, Jinlong Li +3

College of Electronic and Information Engineering, Tongji University, Shanghai, China

Efficient VLM Inference Vision-Language Navigation

Sep 28, 2026cs.CV

Resolution as a First-Class Decision: Task-Conditioned Routing for Efficient Multimodal Large Language Models

The inference efficiency of Multimodal Large Language Models (MLLMs) is severely constrained by massive visual token sequences induced by high-resolution inputs, with computational cost scaling quadratically. Existing approaches primarily focus on downstream token compression, while overlooking a fundamental upstream inefficiency: input resolution is treated as a static, task-agnostic hyperparameter. We propose Task-Conditioned Resolution Routing (TCRR), which formulates visual compression as a task-conditioned decision and employs a lightweight cross-modal router that conditions backbone visual representations on textual semantics via feature-wise modulation and cross-attention to predict the minimal sufficient compression level per query. To support this, we curate a dataset of 500k samples across 12 task categories, labeled via a teacher-oracle pipeline to approximate Pareto-optimal compression scales. Extensive experiments across diverse architectures show that TCRR achieves a superior efficiency frontier, specifically reducing visual FLOPs by 40.9% and latency by 53.7% on Qwen3-VL-8B while preserving competitive performance. Further analysis of scaling behavior confirms that dynamically routing visual compression enables optimal resource allocation without modifying the MLLM backbone.

Zhiqiang Xia, Yang Li, Xinyuan Zhang +5

Xiaomi Corporation, Beijing, China

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

Sep 28, 2026cs.CV

P4Q: Co-designing Token Pruning and Quantization for Vision-Language Model Acceleration

Vision language models have achieved strong performance across a wide range of multimodal applications, yet their substantial computational and memory costs hinder efficient deployment. Visual token pruning and post-training quantization reduce inference overhead along two complementary dimensions, namely sequence length and numerical precision. Existing workflows typically optimize these techniques independently or apply them sequentially. Their distinct optimization objectives leave critical interactions unaddressed and constrain the achievable compression performance. We revisit these designs and present P4Q, a practical co-design framework that jointly optimizes visual token pruning and low-bit quantization for efficient VLM inference. First, P4Q introduces a quantization-aware visual token selection strategy before the LLM. It applies fake quantization to copies of the features produced by the projector and selects visual tokens using statistics computed from these fake-quantized features, thereby conditioning the selector's feature-based decisions on simulated low-bit perturbations. Second, P4Q introduces a pruning-aware quantization calibration strategy. It uses the same selection strategy as pruning to calibrate the quantized model on the retained-token distribution, thereby aligning the calibration process with the pruned execution path used during deployment. By coupling these two components, P4Q achieves substantial inference speedups while maintaining comparable task performance, resulting in a better efficiency-accuracy trade-off than independently optimized pipelines. For instance, on LLaVA-NeXT, P4Q achieves an average end-to-end inference speedup of 2.8x across eight distinct test sets, while retaining higher accuracy than prior compression and quantization methods.

Haizhao Jing, Zhenhao Shang, Haokui Zhang +2

Northwest Polytechnical University · Intellifusion

VLM Quantization Efficient VLM Inference Post-Training Quantization Visual Token Pruning Neural Network Compression

Sep 28, 2026cs.CV

When Text Matters: Design Principles for Visual Token Pruning in Vision-Language Model

Visual token pruning has been widely studied as a practical approach to reducing the computational cost of large vision-language models. However, it struggles to preserve essential visual information, which can lead to substantial performance degradation. In particular, image-based token selection can overlook task-relevant details, while text-guided token selection may fail to capture the text--visual relationships needed for complex reasoning. We find that applying textual guidance too early can limit its ability to identify answer-relevant visual regions, whereas text-to-visual attention becomes more informative at intermediate decoder depths. This finding motivates our training-free method, which separates early vision-guided pruning from deferred text-guided reselection. We first prune visual tokens using vision-encoder attention, retain additional candidates until the decoder midpoint, and then use text-to-visual attention to determine the final visual-token set. Across eight benchmarks and three models, our method outperforms the best-performing baselines by an average of 11.10 and 16.84 percentage points in performance recovery at 80% and 90% pruning, respectively, with comparable or lower LLM-prefill latency than most baselines. The source code is publicly available at https://github.com/kmc3661/DeFT

Minchan Kang, Kyeonghye Park, Seoyoung Cho +2

Korea Advanced Institute of Science and Technology (KAIST) · Hanbat National University

Efficient VLM Inference Visual Token Pruning Visual Token Compression

Sep 28, 2026cs.CV

From Perception to Integration: Revisiting the Internal Dynamics of Reasoning in Vision-Language Models

Vision-language models (VLMs) can answer simple visual questions, but often struggle when one question requires several visual judgments. We study this gap with controlled tasks for feature binding, numerosity, spatial relations, and amodal completion, together with a Composite task that combines them. Matched counterfactual image pairs isolate changes in the visual evidence needed to answer. Across four models, direct answers, hidden-state readouts, and state interventions show that the individual judgments can be made without explicit reasoning and that intervening on the corresponding states can affect the answer. During reasoning, the Composite answer becomes decodable from hidden states and usable from shortened traces, often before the model stops on its own. We train a small detector to predict this readiness and stop reasoning at that point. On MMStar and RealWorldQA, this reduces mean reasoning tokens by 79.1% and 74.5%, while average accuracy rises by 3.13 and 3.30 percentage points, respectively. These findings connect the internal development of answer readiness to a practical rule for allocating reasoning computation.

Rong Yu Xu, Prayag Tiwari, Shaolei Zhang

Shenzhen College of International Education · Halmstad University, Sweden · Renmin University of China

Efficient VLM Inference VLM Interpretability Latent Visual Reasoning VLM Reasoning

Sep 28, 2026cs.CV

ACPruner: Visual Token Pruning as Biased Attention Coverage Maximization in LVLMs

Large Vision-Language Models (LVLMs) face significant computational inefficiencies caused by the large number of visual tokens. Existing visual token pruning methods mainly focus on either retaining individually important tokens or selecting mutually diverse ones. In this work, we revisit visual token pruning from a coverage perspective and formulate it as a biased attention coverage maximization problem. The key idea is to select a compact token subset whose encoder-side outgoing attention can jointly cover the image while assigning higher coverage priority to more informative regions. From this perspective, we propose ACPruner, a training-free visual token pruning framework for efficient LVLM inference. ACPruner first estimates token importance by combining intra-modal saliency and inter-modal relevance, then derives token-wise coverage from attention patterns within the vision encoder, and finally performs greedy selection to maximize the proposed coverage objective. Extensive experiments across multiple LVLM backbones, including LLaVA-1.5-7B/13B, LLaVA-NeXT-7B/13B, Qwen2.5-VL-7B, and LLaVA-OneVision-7B, show that ACPruner consistently achieves strong performance retention while delivering substantial end-to-end inference speedups.

Xu Li, Yuxuan Liang, Yi Zheng +6

College of Computer Science and Artificial Intelligence Fudan University, Shanghai, China

Efficient VLM Inference Large Vision-Language Models Visual Token Pruning

Sep 28, 2026cs.CV

SCOPD: Sparse-Context On-Policy Self-Distillation for Efficient Vision-Language Models

Reasoning vision-language models (VLMs) process images and videos as long sequences of visual tokens, making inference expensive. Training-free token pruning reduces this cost, but aggressive compression can sharply degrade performance, often attributed to irreversible loss of task-relevant visual information. We show that this explanation is incomplete. In a fixed-context Pass@K analysis, repeated sampling from the same pruned visual representation recovers many examples missed by greedy decoding, indicating that useful visual evidence can remain accessible but be used unreliably. We call this the representation-utilization gap. Motivated by this observation, we introduce SCOPD, a sparse-context on-policy self-distillation framework in which a student generates reasoning trajectories from pruned visual tokens while a privileged full-context teacher supervises the same on-policy prefixes. SCOPD requires no ground-truth responses, architectural changes, or additional inference-time computation. We further introduce SCOPD+, which uses a small visual-budget intervention to identify visually sensitive response positions and selectively distill them. At 10% visual-token retention, the Vanilla model retains 86.37% of its unpruned performance across 13 benchmarks. SCOPD raises this to 90.49%, while SCOPD+ further improves it to 92.43%. Across token budgets, benchmarks, and pruning operators, our results show that efficient reasoning depends not only on which visual information survives pruning, but also on how reliably the model learns to use it.

Ahmadreza Jeddi, Enming Zhang, Jasper Gerigk +12

University of Toronto · Vector Institute · Samsung AI Center Toronto +7

Efficient VLM Inference On-Policy Self-Distillation VLM Distillation VLM Reasoning Visual Token Pruning

Sep 27, 2026cs.LG

JET: Justification Evaluation in Transformer

JET uses pretrained language and vision-language models to select among a finite set of answers without additional training. It evaluates candidate likelihoods directly and shares computation across candidates. Experiments on desktop CPUs and consumer GPUs assess decision accuracy and execution cost. Qwen3.6-35B-A3B achieves 87.48% accuracy on the full MMLU test set and 3.69 requests per second on a separately timed MMLU subset. The accuracy-throughput comparison covers model, hardware, and reasoning choices, with Jev as an external reference. Controlled execution experiments show 2.18-2.23-fold speedups from prefix reuse and cache management, and a 30.8% reduction in process time from input preparation optimizations, with unchanged outputs. Optional reasoning has a task-dependent accuracy-throughput trade-off. These results support local decision inference from existing models.

Shenghao Ding

Yet Another AI

Efficient VLM Inference LLM Inference Acceleration Efficient Language Model Inference

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

Vision-Language Models Efficient VLM Inference Visual Token Compression

Sep 26, 2026cs.CV

Fewer Tokens, More Self-Teaching: On-Policy Self-Distillation for Extreme Visual Token Reduction

Visual token reduction is an effective way to accelerate multimodal large language models (MLLMs), but performance deteriorates rapidly under extremely low token budgets. Existing work has explored both visual-token selection and training-based adaptation to reduced visual inputs. We take a step further by asking how a heavily compressed MLLM should learn from the states induced by its own generations. This setting naturally calls for on-policy self-distillation: a heavily compressed model is supervised on the states induced by its own generations, while its full-token counterpart serves as an information-rich teacher. Based on this insight, we propose LT-OPD, a training framework for extreme visual-token reduction. The student rolls out responses with only a small fraction of visual tokens, and a frozen full-token copy of the same MLLM provides distributional supervision along these student-generated trajectories. To stabilize on-policy learning when visual evidence is severely limited, we further introduce a budget-level curriculum that progressively decreases the token budget during training. Across nine benchmarks on Qwen3.5-4B, LT-OPD raises average retained performance under 5% visual-token retention from 68.6% to 82.3%, outperforming training-free, training-based, and reinforcement-learning baselines at the same budget. The gains transfer consistently to Qwen3.5-9B, GLM-4.6V-9B, and LLaVA-OV-1.5-4B. LT-OPD also reduces KV-cache usage by 85.2% and prefill FLOPs by 85.4% without additional inference overhead, demonstrating that on-policy learning can substantially recover capabilities lost to extreme visual-token reduction.

Junxian Li, Ruixuan Yang, Tianao Zhang +3

Shanghai Jiao Tong University · Xi’an Jiaotong University · University of Cambridge +1

Efficient VLM Inference VLM Distillation Large Vision-Language Models On-Policy Distillation Visual Token Compression

Sep 24, 2026cs.AI

Jev-Mobile: Jev as an Executor for Mobile GUI Agents

Vision-language models (VLMs) have become a common foundation for autonomous mobile GUI agents, but most existing systems rely on the VLM for both planning and action grounding at nearly every interaction step, leading to substantial latency and model-serving cost. We introduce Jev-Mobile, which shifts this paradigm to low-frequency VLM planning and high-frequency lightweight execution: the VLM specifies local goals, the accessibility tree defines a structured executable action space, and Jev, a fast typed decision model, repeatedly selects actions within this space. This design allows multiple GUI actions to be executed under a single VLM decision, reducing expensive VLM inference while preserving adaptive interaction. On the full AndroidWorld task suite, Jev-Mobile achieves 79% task success, compared with 78% for SeeAct-V and 84% for a Step-wise VLM baseline. Among successful trajectories, it reduces mean end-to-end execution time by 32.7% and mean model API cost by 73.4% relative to Step-wise VLM. These results show that decoupling high-level VLM reasoning from low-level action execution can substantially improve mobile GUI agent efficiency while maintaining competitive task performance.

Linghua Zhang

Efficient VLM Inference Mobile GUI Agents