arXiv Computer Vision August 27, 2026 By Haobo Xiong, Shaobo Liu, Kai Liu, Chongyang Ding

CrossMambaTuning: Synergistic Spatial and Cross-Layer Adaptation for Machine Vision Compression

Read the original on arXiv Computer Vision →

CrossMambaTuning is a new framework that adapts pretrained learned image compression models to machine vision tasks with minimal retraining. It combines State Space Models with cross‑layer interaction, featuring a Mamba adapter that uses task‑specific prompts and multi‑scale branching, and a Scale‑Invariant Cross‑Layer Adapter (SICA) that shares parameters across scales to reduce redundancy. Experiments show that this approach achieves state‑of‑the‑art performance while cutting parameter overhead by 72% compared to existing methods.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

Hugging Face Trending Papers

Aug 26

CrossMambaTuning: Synergistic Spatial and Cross-Layer Adaptation for Machine Vision Compression

To reduce deployment cost and retraining overhead, adapting pretrained learned image compression (LIC) models to downstream machine vision tasks has attracted growing attention. However, existing meth...

fine-tuning benchmarks

arXiv AI

Jun 2

On the Limits of Token Reduction for Efficient Unified Vision Language Training

arXiv:2606. 01503v1 Announce Type: cross Abstract: Unified vision-language models (VLMs) integrate visual understanding and visual generation within a single autoregressive backbone, but their joint training is computationally expensive and largely overlooked from an efficiency perspective.

By Siyi Chen, Weiming Zhuang, Jingtao Li, Lingjuan Lv

llms multimodal

arXiv Computer Vision

Sep 22

LiAuto-MindViT: A Hybrid Vision Backbone with Adaptive Bidirectional Mamba

arXiv:2609.24337v1 Announce Type: new Abstract: While Mamba-based models have shown strong potential for long sequence modeling, adapting them to vision is challenging due to the requirement of local...

By Lifu Mu, Shuai Chen, Wen Zheng, Haoyi Sun, Xueyang Fu, Pengfei Yu, Ning Mao, Tao Wei, Zhou Pan

llms computer-vision benchmarks safety

arXiv AI

Jul 7

Fusion: A Framework for Unified Sequential Token AdaptatIon in VisiOn TraNsformers

arXiv:2607. 02612v1 Announce Type: cross Abstract: Vision Transformers achieve strong image classification accuracy but process all image regions with nearly the same computation, even when many regions are redundant or uninformative.

By Aravind Pradeep, Samira Nazari, Mahdi Taheri, Christian Herglotz

llms computer-vision efficiency benchmarks

arXiv Computer Vision

Oct 19

ReVision: Reprogramming Image Generators for Dense Visual Perception

arXiv:2610.23533v1 Announce Type: new Abstract: Recent image generators exhibit rich visual priors that can support dense visual perception. However, existing approaches typically encode multiple per...

By Yitong Yang, Yinglin Wang, Xuexin Liu, Wei Chen, Zhaoyang Xu, Jiaxin Lin, Rui Chen, Shuting He, Kai Wang

computer-vision multimodal benchmarks

arXiv Computer Vision

Sep 11

LoopVAE: Recurrent Depth Across Scales for Visual Tokenization

LoopVAE introduces a recurrent depth architecture that reuses a scale‑ and loop‑conditioned core across different spatial scales while keeping resolution‑changing transitions separate. The four‑block core applies 28 block operations per encoder or decoder, enabling a 29M‑parameter convolutional model to achieve 0.28 rFID and 32.54 dB PSNR on ImageNet‑256 with roughly 65% fewer parameters than comparable VAEs. Experiments with both convolutional and Transformer operators, as well as ablations on parameter sharing, demonstrate competitive image quality metrics and reveal how targeted loop interventions and truncation affect reconstruction quality and computational trade‑offs.

By Zhiying Lu

llms nlp benchmarks safety

One email a morning, machine-written

Related stories

CrossMambaTuning: Synergistic Spatial and Cross-Layer Adaptation for Machine Vision Compression

On the Limits of Token Reduction for Efficient Unified Vision Language Training

LiAuto-MindViT: A Hybrid Vision Backbone with Adaptive Bidirectional Mamba

Fusion: A Framework for Unified Sequential Token AdaptatIon in VisiOn TraNsformers

ReVision: Reprogramming Image Generators for Dense Visual Perception

LoopVAE: Recurrent Depth Across Scales for Visual Tokenization