[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85500-en":3,"doc-seo-85500-105":29,"detail-sidebar-cat-0-en-105":90},{"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":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},85500,962075006959,"Anda","https://ap-avatar.wpscdn.com/avatar/e0002397efbe92a78e?_k=1776741047341049297",8,"Research & Report","TimeSAE Causal Sparse Decoding for Faithful Explanations of Black-Box Time Series Models","As black-box and pretrained models expand time-series applications, interpreting and trusting their forecasts becomes critical in domains such as finance, healthcare, and energy management. Existing explanation techniques often remain limited to in-distribution settings and fail under distribution shifts, reducing real-world reliability. This work introduces TimeSAE, a causal sparse autoencoder framework combining Sparse Autoencoders and causality. Extensive evaluations on synthetic and real-world datasets compare favorably with strong baselines, yielding more faithful and robust explanations supported by both quantitative and qualitative evidence.","TimeSAE: Causal Sparse Decoding for Faithful Explanations of Black-Box Time Series Models  \nKhalid Oublal* 1 Quentin Bouniot 1 2 3 4 5 Qi Gan 1 Stephan Clémençon 1 Zeynep Akata 2 3 4 5  \narXiv :2601 .09776v2 [ cs .LG] 11 Jul 2026  \nAbstract  \nAs black box models and pretrained models gain traction in time series applications, understanding and explaining their predictions becomes increasingly vital, especially in high-stakes domains where interpretability and trust are essential. However, most of the existing methods involve only in-distribution explanation, and do not generalize outside the training support, which requires the learning capability of generalization. In this work, we aim to provide a framework to explain blackbox models for time series data through the dual lenses of Sparse Autoencoders (SAEs) and causality. We show that many current explanation methods are sensitive to distributional shifts, limiting their effectiveness in real-world scenarios. Building on the concept of Sparse Autoencoder, we introduce TimeSAE , a framework for black-box model explanation. We conduct extensive evaluations of TimeSAE on both synthetic and realworld time series datasets, comparing it to leading baselines. The results, supported by both quantitative metrics and qualitative insights, show that TimeSAE delivers more faithful and robust explanations. Our code and dataset are available in an easy-to-use library TimeSAE-Lib: [https://](https://)[ ](https://)[oublalkhalid.github.io/TimeSAE/](oublalkhalid.github.io/TimeSAE/) .  \n1. Introduction  \nThe rise of black box models such as large foundation models has revolutionized various fields, including time series analysis, with applications in finance (Bento et al., 2021), healthcare (Kaushik et al., 2020), and environmental science (Adebayo et al., 2021) . These networks often make  \n1LTCI, Télécom Paris, Institut Polytechnique de Paris 2Technical University of Munich 3Helmholtz Munich 4Munich Center for Machine Learning 5Munich Data Science Institute. Correspondence to: Khalid Oublal \u003C[khalid.oublal@ip-paris.fr](khalid.oublal@ip-paris.fr)>, Quentin Bouniot \u003C[quentin.bouniot@telecom-paris.fr](quentin.bouniot@telecom-paris.fr)>.  \nProceedings of the 43 rd International Conference on Machine Learning, Seoul, South Korea. PMLR 306, 2026 . Copyright 2026 by the author(s) .  \ncritical decisions, especially in sensitive domains where decisions are based on forecasting outcomes, such as managing grid stability in Energy (Eid et al., 2016), and Healthcare (Dairi et al., 2021), yet the underlying decision-making process is difficult to interpret due to the black-box nature of the models. This opacity has motivated the rise of explainable AI (XAI) techniques to provide human-understandable explanations for model decisions. While XAI has been predominantly applied in image classification, it is extending into other fields, such as audio and time series (Parekh et al., 2022 ; Queen et al., 2023) .  \nCurrent methods in enhancing explainability for time series primarily identify key signal locations (sub-instance) affecting model predictions. For instance, Shi et al. (2023) uses LIME (Ribeiro et al., 2016) to explain water level prediction models. Additionally, perturbation methods like Dynamask (Crabbé & Van Der Schaar, 2021) and Extrmask (Enguehard, 2023) modify less critical features to evaluate their impact but often struggle with feature interdependencies and generalization. Despite their insights, these techniques face challenges with Out-of-Distribution (OOD) samples, affecting the faithfulness of explanations (Queen et al., 2023) .  \nExplaning time series black-box models requires the ability to generalize beyond the training distribution, which is essential for the robust deployment of explanatory algorithms in real-world scenarios. In addressing the extrapolation of explanation, Queen et al. (2023) retrain a white-box model for consistency, though this depends on knowing the model’s struct","cbCaigiYbqrtNfdW","https://ap.wps.com/l/cbCaigiYbqrtNfdW","pdf",1707406,1,36,"English","en",105,"# Introduction\n## Background and motivation\n## Prior work in time-series explainability\n## Faithful explanations and definition\n## Time Sparse Autoencoder (TimeSAE)\n## Framework overview","[{\"question\":\"Why is explainability especially important for black-box time series models?\",\"answer\":\"Black-box time-series models drive critical decisions where forecasting outcomes are used, but their internal reasoning is difficult to interpret. Explainable AI techniques address the need for human-understandable explanations in such settings.\"},{\"question\":\"What limitation do many existing time-series explanation methods have?\",\"answer\":\"Many methods focus on in-distribution explanations and do not generalize outside the training support. As a result, explanations can become sensitive to distributional shifts, weakening their effectiveness in real-world scenarios.\"},{\"question\":\"How does TimeSAE improve faithfulness and robustness of explanations?\",\"answer\":\"TimeSAE combines Sparse Autoencoders with causality to form a framework that targets faithful explanations. It also evaluates performance on synthetic and real-world time-series datasets and reports stronger results than leading baselines using both quantitative metrics and qualitative insights.\"}]",1784204034,91,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"timesae-causal-sparse-decoding-for-faithful-explanations-of-black-box-time-series-models","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/timesae-causal-sparse-decoding-for-faithful-explanations-of-black-box-time-series-models/85500/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"Why is explainability especially important for black-box time series models?","Question",{"text":74,"@type":75},"Black-box time-series models drive critical decisions where forecasting outcomes are used, but their internal reasoning is difficult to interpret. Explainable AI techniques address the need for human-understandable explanations in such settings.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"What limitation do many existing time-series explanation methods have?",{"text":79,"@type":75},"Many methods focus on in-distribution explanations and do not generalize outside the training support. As a result, explanations can become sensitive to distributional shifts, weakening their effectiveness in real-world scenarios.",{"name":81,"@type":72,"acceptedAnswer":82},"How does TimeSAE improve faithfulness and robustness of explanations?",{"text":83,"@type":75},"TimeSAE combines Sparse Autoencoders with causality to form a framework that targets faithful explanations. It also evaluates performance on synthetic and real-world time-series datasets and reports stronger results than leading baselines using both quantitative metrics and qualitative insights.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":105,"slug":137},19,"General","general"]