[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84462-en":3,"doc-seo-84462-105":29,"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":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},84462,687197100911,"Himbo","https://ap-avatar.wpscdn.com/avatar/a000239b6f1da00475?x-image-process=image/resize,m_fixed,w_180,h_180&k=1782698725881665579",8,"Research & Report","CoGenCast Coupled Autoregressive Flow Generative Framework for Time Series Forecasting","CoGenCast treats time series forecasting as a generative task requiring both semantic understanding of contextual conditions and stochastic modeling of continuous temporal dynamics. The proposed hybrid framework couples pre-trained decoder-only LLMs with a flow-matching mechanism, reformulating LLMs into an encoder–decoder forecasting backbone via attention-topology changes. A flow-matching denoising component models temporal evolution as continuous stochastic dynamics conditioned on the autoregressive representation, enabling multimodal forecasting and cross-domain unified training. Experiments on multiple benchmarks demonstrate competitive results against prior baselines.","CoGenCast: A Coupled Autoregressive-Flow Generative Framework  \nfor Time Series Forecasting  \nMingyue Cheng 1 Yaguo Liu 1 Daoyu Wang 1 Xiaoyu Tao 1 Qi Liu* 1  \narXiv :2602 .03564v2 [ cs .LG] 11 Jul 2026  \nAbstract  \nTime series forecasting can be viewed as a generative problem that requires both semantic understanding over contextual conditions and stochastic modeling of continuous temporal dynamics.  \nExisting approaches typically rely on either autoregressive large language models (LLMs) for semantic context modeling or diffusion-like models for continuous probabilistic generation. However, neither method alone can adequately model both aspects simultaneously. In this work, we propose CoGenCast, a hybrid generative framework that couples pre-trained LLMs with flowmatching mechanism for effective time series forecasting. Specifically, we reconfigure pre-trained decoder-only LLMs into a native forecasting encoder–decoder backbone by modifying only the attention topology, enabling bidirectional context encoding and causal representation generation. Building on this, a flow-matching mechanism is further integrated to model temporal evolution, capturing continuous stochastic dynamics conditioned on the autoregressively generated representation. Notably, CoGenCast naturally supports multimodal forecasting and crossdomain unified training. Extensive experiments on multiple benchmarks show that CoGenCast achieves competitive performance compared to previous baselines. Code is available at [https:](https:)//[github.com/liuyaguo/_CoGenCast](github.com/liuyaguo/_CoGenCast).  \n1. Introduction  \nTime series forecasting supports a wide range of real-world decision-making processes, including energy (Zhou et al., 2024), finance  (Feng et al., 2019) and healthcare (Qiu et al.,  \n1 State Key Laboratory of Cognitive Intelligence, University of Science and Technology of China, Hefei, China. Correspondence to: Qi Liu \u003C [qiliuql@ustc.edu.cn](qiliuql@ustc.edu.cn) >.  \nProceedings of the 43 rd International Conference on Machine Learning, Seoul, South Korea. PMLR 306, 2026 . Copyright 2026 by the author(s) .  \n2024) . The core philosophy of forecasting (Liu et al., 2023) is to predict future values conditioned on historical observations, while future evolution is often full of complexity and uncertainty. To handle this, previous works typically implement a simple regression mapping (Nie et al., 2022) . However, recently, an increasing number of methods employ more powerful mapping functions, such as LLMs (Niu et al., 2025 ; Jiang et al., 2025) and diffusion-like (Guo et al., 2025) models. These methods reformulate forecasting as a conditional generative problem (Liu et al., 2024d) .  \nCurrent generative forecasting methods can be divided into two main categories: LLM-based methods and diffusionlike models. First, LLM-based methods leverage the strong semantic understanding and generation capabilities of LLMs. These can be further branched into tuning-based methods (Jin et al., 2024a ; Xie et al., 2025), which finetune LLMs to align them with numerical time series modeling, and training-free methods (Liu et al., 2024b ; Cheng et al., 2026), which keep LLMs frozen and rely on prompting or reasoning mechanisms to perform forecasting. Second, diffusion-like models focus on modeling the probabilistic evolution of future time series in continuous-valued spaces. These approaches effectively capture uncertainty in future temporal trajectories. Representative works such as NsDiff (Ye et al., 2025) and TSFlow (Kollovieh et al., 2025) learn continuous stochastic temporal dynamics, enabling high-quality probabilistic generation (Tashiro et al., 2021) .  \nAlthough the above generative forecasting methods achieve promising results, we argue that an ideal forecasting approach should simultaneously possess dual capabilities: semantic understanding over contextual conditions and stochastic modeling of continuous temporal dynamics. Taking electricity load forecasti","cbCaikXJn4nwOaK5","https://ap.wps.com/l/cbCaikXJn4nwOaK5","pdf",2885073,1,25,"English","en",105,"# Abstract\n# Introduction\n## Forecasting as a generative problem\n## Existing approaches: LLM-based vs diffusion-like\n## Motivation: dual semantic and stochastic capabilities\n## CoGenCast framework\n## Main contributions","[{\"question\":\"What problem does CoGenCast address in time series forecasting?\",\"answer\":\"CoGenCast addresses forecasting as a generative task that needs both semantic understanding of contextual conditions and stochastic modeling of continuous temporal dynamics.\"},{\"question\":\"How does CoGenCast use pre-trained LLMs in its framework?\",\"answer\":\"It reconfigures pre-trained decoder-only LLMs into a forecasting encoder–decoder backbone by modifying only the attention topology, enabling bidirectional context encoding and causal autoregressive representation generation.\"},{\"question\":\"What role does the flow-matching mechanism play in CoGenCast?\",\"answer\":\"The integrated flow-matching denoising decoder models temporal evolution, capturing continuous stochastic dynamics conditioned on the autoregressively generated representation and supporting efficient generation for multimodal forecasting.\"}]",1784195783,63,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":27},"cogencast-coupled-autoregressive-flow-generative-framework-for-time-series-forecasting","",{"@graph":35,"@context":85},[36,53,68],{"@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/cogencast-coupled-autoregressive-flow-generative-framework-for-time-series-forecasting/84462/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-22","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 CoGenCast address in time series forecasting?","Question",{"text":75,"@type":76},"CoGenCast addresses forecasting as a generative task that needs both semantic understanding of contextual conditions and stochastic modeling of continuous temporal dynamics.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does CoGenCast use pre-trained LLMs in its framework?",{"text":80,"@type":76},"It reconfigures pre-trained decoder-only LLMs into a forecasting encoder–decoder backbone by modifying only the attention topology, enabling bidirectional context encoding and causal autoregressive representation generation.",{"name":82,"@type":73,"acceptedAnswer":83},"What role does the flow-matching mechanism play in CoGenCast?",{"text":84,"@type":76},"The integrated flow-matching denoising decoder models temporal evolution, capturing continuous stochastic dynamics conditioned on the autoregressively generated representation and supporting efficient generation for multimodal forecasting.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]