[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82864-en":3,"doc-seo-82864-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},82864,8796095462418,"Noah","https://ap-avatar.wpscdn.com/avatar/80000253c1241d02b47?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778826106357471780",8,"Research & Report","Discovering Shared Interpretable Operations in Image Compression Autoencoders","Image compression autoencoders often improve rate–distortion performance by increasing model size and hiding internal knowledge in black-box nonlinear operations. This paper analyzes bias-free autoencoders using Jacobian-based techniques to detect universal, shared behaviors inside the learned processing. When consistent input-independent or input-aligned filtering patterns exist, they can be aggregated across images and reused to build low-complexity compression models. The approach aims to preserve reconstruction quality while reducing model operations.","Discovering shared interpretable operations in image compression autoencoders  \nCaroline Mazini Rodrigues, Nicolas Keriven, and Thomas Maugey Univ. Rennes, Inria, CNRS, IRISA, Rennes, France  \narXiv :2607 .04839v1 [ ee ss .IV] 6 Jul 2026  \nAbstract—With the increasing adoption of deep learning for applications such as image compression, improvements in the rate-distortion trade-off have been achieved at the cost of increasingly larger and more opaque “black-box” models. Autoencoders are among the most widely used architectures for this task; however, without a clear understanding of their internal behavior, these models tend to grow in complexity to achieve more performance gains. In this paper, we investigate whether universal behaviors can be detected from the internal operations of bias-free autoencoders through Jacobian analysis. If such behaviors exist, they may be extracted to design low-complexity image compression models inspired by high-complexity deep learning architectures.  \nIndex Terms—Interpretability, Image compression, Explainable Artificial Intelligence, Jacobian analysis, Frugality  \nI. INTRODUCTION  \nAutoencoders are widely used for image compression tasks, whose main objective is to reduce the amount of transmitted information (rate) while preserving a low reconstruction error (distortion) at the receiver side [1],[2] . These models have progressively replaced handcrafted transformations, such as the Discrete Cosine Transform (DCT), because they can achieve better rate–distortion trade-offs [3] . However, improvements in compression performance are often obtained by increasing the size and complexity of the models. This can limit their applicability in real-world scenarios with constrained hardware and may also lead to higher energy consumption [4], [5], [6] .  \nCurrently, larger deep learning models achieve high performance by adapting to their input data. However, the knowledge learned during training remains hidden within these models, which are often treated as “black boxes” due to their complex nonlinear operations and large scale [7] . If this hidden knowledge could be better understood, we may use it more directly, reducing the need for some of the original model operations and potentially improving efficiency.  \nSome studies, such as Mohan et al. [8] and Kadkhodaie et al. [9], propose interpreting autoencoder models as a set of input-adapted linear operations, similar to handcrafted linear filters. However, these explanations are mainly local, since they focus on the model’s adaptation to individual images [10],[11], [12] . This contrasts with global explanations, which aim to identify universal mechanisms consistently used by the model across different input data [13], [14], [15] .  \nInspired by the idea of describing a model as a set of inputadapted linear filters, we investigate the universality of these  \nThe authors acknowledge fundings of France 2030, PEPR IA, ANR-23-PEIA-0008 and European Union ERC-2024-STG-101163069 MALAGA.  \nfilters across different images in the compression task. Figure 1 shows the similarities of image-adapted filters extracted from a compression model for two different images and raises the questions: does the model exhibit a general behavior across images, and at what point does it become image-specific? Our intuition is that, if some of these filters are common or similar across images, they could be used to develop low-complexity, deep-learning-inspired compression tools.  \nFig. 1: Filtering operations by pixel in a compression autoencoder. In the first row, we show pixel positions in two images filtered by the model. We present a schematic of the filters’ construction in Figure 2 . The same three positions are used for all images. We show the three most important filters for preserving reconstruction quality, as identified through Jacobian analysis, and observe consistent behavior across both images for the most influential features. The white image represents our pr","cbCaieqrYVIogOZc","https://ap.wps.com/l/cbCaieqrYVIogOZc","pdf",4088531,4,1,6,"English","en",105,"# Introduction\n## Motivation and background\n## Jacobian-based interpretability approach\n# Do compression models learn universality?\n## Bias-free Jacobian analysis\n## Rate–distortion objective in autoencoders","[{\"question\":\"What problem does the paper address in image compression autoencoders?\",\"answer\":\"It addresses the trade-off where better compression performance often comes from larger, more opaque “black-box” models that are hard to interpret internally.\"},{\"question\":\"How does the paper use Jacobian analysis to interpret autoencoders?\",\"answer\":\"It analyzes bias-free autoencoders via per-image Jacobians, extracting image-adaptive linear filtering operations from the Jacobian rows.\"},{\"question\":\"How can shared behaviors across images help compression model design?\",\"answer\":\"If common or similar filtering operations persist across inputs, they can be aggregated and used to construct low-complexity compression models inspired by high-complexity deep architectures.\"}]",1784183539,15,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"discovering-shared-interpretable-operations-in-image-compression-autoencoders","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/discovering-shared-interpretable-operations-in-image-compression-autoencoders/82864/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the paper address in image compression autoencoders?","Question",{"text":75,"@type":76},"It addresses the trade-off where better compression performance often comes from larger, more opaque “black-box” models that are hard to interpret internally.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the paper use Jacobian analysis to interpret autoencoders?",{"text":80,"@type":76},"It analyzes bias-free autoencoders via per-image Jacobians, extracting image-adaptive linear filtering operations from the Jacobian rows.",{"name":82,"@type":73,"acceptedAnswer":83},"How can shared behaviors across images help compression model design?",{"text":84,"@type":76},"If common or similar filtering operations persist across inputs, they can be aggregated and used to construct low-complexity compression models inspired by high-complexity deep architectures.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,114,119,122,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]