{"about":{"non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","site":"https://codewithpapers.app","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/xpatch-dual-stream-time-series-forecasting","title":"xPatch: Dual-Stream Time Series Forecasting with Exponential Seasonal-Trend Decomposition","arxiv_id":"2412.17323","date":"2024-12-23","proceeding":null,"authors":["Artyom Stitsyuk","Jaesik Choi"],"abstract":"In recent years, the application of transformer-based models in time-series forecasting has received significant attention. While often demonstrating promising results, the transformer architecture encounters challenges in fully exploiting the temporal relations within time series data due to its attention mechanism. In this work, we design eXponential Patch (xPatch for short), a novel dual-stream architecture that utilizes exponential decomposition. Inspired by the classical exponential smoothing approaches, xPatch introduces the innovative seasonal-trend exponential decomposition module. Additionally, we propose a dual-flow architecture that consists of an MLP-based linear stream and a CNN-based non-linear stream. This model investigates the benefits of employing patching and channel-independence techniques within a non-transformer model. Finally, we develop a robust arctangent loss function and a sigmoid learning rate adjustment scheme, which prevent overfitting and boost forecasting performance. The code is available at the following repository: https://github.com/stitsyuk/xPatch.","url_abs":"https://arxiv.org/abs/2412.17323v2","url_pdf":"https://arxiv.org/pdf/2412.17323v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"xpatch-dual-stream-time-series-forecasting","repo_url":"https://github.com/stitsyuk/xpatch","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"multivariate-time-series-forecasting","task_name":"Multivariate Time Series Forecasting"},{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series-forecasting","task_name":"Time Series Forecasting"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"patching","method_name":"Patching"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/time-series-forecasting-on-etth1-192-1","task":"Time Series Forecasting","dataset":"ETTh1 (192) Multivariate","model":"xPatch","rank_in_archive_order":2,"of":17,"metrics":{"MAE":"0.395","MSE":"0.376"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-etth1-336-1","task":"Time Series Forecasting","dataset":"ETTh1 (336) Multivariate","model":"xPatch","rank_in_archive_order":2,"of":72,"metrics":{"MAE":"0.415","MSE":"0.391"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-etth1-720-1","task":"Time Series Forecasting","dataset":"ETTh1 (720) Multivariate","model":"xPatch","rank_in_archive_order":6,"of":22,"metrics":{"MAE":"0.459","MSE":"0.442"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-etth1-96-1","task":"Time Series Forecasting","dataset":"ETTh1 (96) Multivariate","model":"xPatch","rank_in_archive_order":3,"of":15,"metrics":{"MAE":"0.379","MSE":"0.354"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-etth2-192-1","task":"Time Series Forecasting","dataset":"ETTh2 (192) Multivariate","model":"xPatch","rank_in_archive_order":2,"of":16,"metrics":{"MAE":"0.330","MSE":"0.275"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-etth2-336-1","task":"Time Series Forecasting","dataset":"ETTh2 (336) Multivariate","model":"xPatch","rank_in_archive_order":1,"of":20,"metrics":{"MAE":"0.360","MSE":"0.312"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-etth2-720-1","task":"Time Series Forecasting","dataset":"ETTh2 (720) Multivariate","model":"xPatch","rank_in_archive_order":4,"of":20,"metrics":{"MAE":"0.418","MSE":"0.384"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-etth2-96-1","task":"Time Series Forecasting","dataset":"ETTh2 (96) Multivariate","model":"xPatch","rank_in_archive_order":2,"of":16,"metrics":{"MAE":"0.297","MSE":"0.226"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-ettm1-192-1","task":"Time Series Forecasting","dataset":"ETTm1 (192) Multivariate","model":"xPatch","rank_in_archive_order":1,"of":9,"metrics":{"Accuracy":"0.355","MSE":"0.315"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-ettm1-336-1","task":"Time Series Forecasting","dataset":"ETTm1 (336) Multivariate","model":"xPatch","rank_in_archive_order":2,"of":8,"metrics":{"MAE":"0.376","MSE":"0.355"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-ettm1-720-1","task":"Time Series Forecasting","dataset":"ETTm1 (720) Multivariate","model":"xPatch","rank_in_archive_order":4,"of":8,"metrics":{"MAE":"0.411","MSE":"0.419"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-ettm1-96-1","task":"Time Series Forecasting","dataset":"ETTm1 (96) Multivariate","model":"xPatch","rank_in_archive_order":1,"of":9,"metrics":{"MAE":"0.330","MSE":"0.275"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-ettm2-192-1","task":"Time Series Forecasting","dataset":"ETTm2 (192) Multivariate","model":"xPatch","rank_in_archive_order":1,"of":9,"metrics":{"MAE":"0.280","MSE":"0.213"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-ettm2-336-1","task":"Time Series Forecasting","dataset":"ETTm2 (336) Multivariate","model":"xPatch","rank_in_archive_order":2,"of":8,"metrics":{"MAE":"0.315","MSE":"0.264"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-ettm2-720-1","task":"Time Series Forecasting","dataset":"ETTm2 (720) Multivariate","model":"xPatch","rank_in_archive_order":1,"of":8,"metrics":{"MAE":"0.363","MSE":"0.338"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-ettm2-96-1","task":"Time Series Forecasting","dataset":"ETTm2 (96) Multivariate","model":"xPatch","rank_in_archive_order":1,"of":9,"metrics":{"MAE":"0.240","MSE":"0.153"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-electricity-192","task":"Time Series Forecasting","dataset":"Electricity (192)","model":"xPatch","rank_in_archive_order":1,"of":8,"metrics":{"MAE":"0.232","MSE":"0.140"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-electricity-336","task":"Time Series Forecasting","dataset":"Electricity (336)","model":"xPatch","rank_in_archive_order":1,"of":9,"metrics":{"MAE":"0.249","MSE":"0.156"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-electricity-720","task":"Time Series Forecasting","dataset":"Electricity (720)","model":"xPatch","rank_in_archive_order":4,"of":8,"metrics":{"MAE":"0.281","MSE":"0.190"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-electricity-96","task":"Time Series Forecasting","dataset":"Electricity (96)","model":"xPatch","rank_in_archive_order":1,"of":11,"metrics":{"MAE":"0.217","MSE":"0.126"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-exchange-192-1","task":"Time Series Forecasting","dataset":"Exchange (192)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.298","MSE":"0.178"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-exchange-336-1","task":"Time Series Forecasting","dataset":"Exchange (336)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.418","MSE":"0.339"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-exchange-720-1","task":"Time Series Forecasting","dataset":"Exchange (720)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.701","MSE":"0.867"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-exchange-96-1","task":"Time Series Forecasting","dataset":"Exchange (96)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.197","MSE":"0.081"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-illness-24-1","task":"Time Series Forecasting","dataset":"Illness (24)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.638","MSE":"1.188"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-illness-36-1","task":"Time Series Forecasting","dataset":"Illness (36)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.653","MSE":"1.226"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-illness-48-1","task":"Time Series Forecasting","dataset":"Illness (48)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.686","MSE":"1.254"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-illness-60-1","task":"Time Series Forecasting","dataset":"Illness (60)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy":"0.773","MSE":"1.455"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-solar-192-1","task":"Time Series Forecasting","dataset":"Solar (192)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.216","MSE":"0.193"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-solar-336-1","task":"Time Series Forecasting","dataset":"Solar (336)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.224","MSE":"0.196"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-solar-720-1","task":"Time Series Forecasting","dataset":"Solar (720)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.219","MSE":"0.212"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-solar-96-1","task":"Time Series Forecasting","dataset":"Solar (96)","model":"xPatch","rank_in_archive_order":1,"of":1,"metrics":{"MAE":"0.197","MSE":"0.173"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-traffic-192","task":"Time Series Forecasting","dataset":"Traffic (192)","model":"xPatch","rank_in_archive_order":2,"of":3,"metrics":{"MAE":"0.241","MSE":"0.377"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-traffic-336","task":"Time Series Forecasting","dataset":"Traffic (336)","model":"xPatch","rank_in_archive_order":3,"of":3,"metrics":{"MAE":"0.243","MSE":"0.388"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-traffic-720","task":"Time Series Forecasting","dataset":"Traffic (720)","model":"xPatch","rank_in_archive_order":3,"of":3,"metrics":{"MAE":"0.273","MSE":"0.437"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-traffic-96","task":"Time Series Forecasting","dataset":"Traffic (96)","model":"xPatch","rank_in_archive_order":3,"of":3,"metrics":{"MAE":"0.233","MSE":"0.364"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-weather-192","task":"Time Series Forecasting","dataset":"Weather (192)","model":"xPatch","rank_in_archive_order":5,"of":13,"metrics":{"MAE":"0.227","MSE":"0.189"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-weather-336","task":"Time Series Forecasting","dataset":"Weather (336)","model":"xPatch","rank_in_archive_order":1,"of":11,"metrics":{"MAE":"0.260","MSE":"0.218"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-weather-720","task":"Time Series Forecasting","dataset":"Weather (720)","model":"xPatch","rank_in_archive_order":1,"of":11,"metrics":{"MAE":"0.315","MSE":"0.291"},"uses_additional_data":false},{"leaderboard":"/sota/time-series-forecasting-on-weather-96","task":"Time Series Forecasting","dataset":"Weather (96)","model":"xPatch","rank_in_archive_order":7,"of":12,"metrics":{"MAE":"0.185","MSE":"0.146"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2412.17323","atlas_url":"https://app.syntology.ai/?focus=2412.17323","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2412.17323"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2412.17323\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2412.17323"},"call":"get_concepts_for_paper(arxiv_id=\"2412.17323\")"}],"by_repo":[{"repo":"stitsyuk/xpatch","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"a0c5cc5ee60ab3c3ed80ed5513d26ab66e0df8228ce2e614f11d4f65c0b9bab5","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}