{"about":{"site":"https://codewithpapers.app","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.","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"},"url":"/paper/multi-scale-finetuning-for-encoder-based-time","title":"Multi-Scale Finetuning for Encoder-based Time Series Foundation Models","arxiv_id":"2506.14087","date":"2025-06-17","proceeding":null,"authors":["Zhongzheng Qiao","Chenghao Liu","Yiming Zhang","Ming Jin","Quang Pham","Qingsong Wen","P. N. Suganthan","Xudong Jiang","Savitha Ramasamy"],"abstract":"Time series foundation models (TSFMs) demonstrate impressive zero-shot performance for time series forecasting. However, an important yet underexplored challenge is how to effectively finetune TSFMs on specific downstream tasks. While naive finetuning can yield performance gains, we argue that it falls short of fully leveraging TSFMs' capabilities, often resulting in overfitting and suboptimal performance. Given the diverse temporal patterns across sampling scales and the inherent multi-scale forecasting capabilities of TSFMs, we adopt a causal perspective to analyze finetuning process, through which we highlight the critical importance of explicitly modeling multiple scales and reveal the shortcomings of naive approaches. Focusing on \\textit{encoder-based} TSFMs, we propose \\textbf{M}ulti\\textbf{\\textsc{s}}cale \\textbf{\\textsc{f}}ine\\textbf{\\textsc{t}}uning (\\textbf{MSFT}), a simple yet general framework that explicitly integrates multi-scale modeling into the finetuning process. Experimental results on three different backbones (\\moirai, \\moment\\ and \\units) demonstrate that TSFMs finetuned with MSFT not only outperform naive and typical parameter efficient finetuning methods but also surpass state-of-the-art deep learning methods.","url_abs":"https://arxiv.org/abs/2506.14087v1","url_pdf":"https://arxiv.org/pdf/2506.14087v1.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":[],"tasks":[{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series-forecasting","task_name":"Time Series Forecasting"}],"methods":[{"method_slug":"adopt","method_name":"ADOPT"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2506.14087","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2506.14087"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/zqiao11/MSFT","reach":{"status":"ok","spdx":"NOASSERTION"}}],"summary":{"unverified":13},"by_repo_kind":{"found_in_text":{"samples":13,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":13,"samples":[{"code_sha256_prefix":"dd639b004340cd07","entry":"beta_binomial_sampler","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/common/sampler.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/common/sampler.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dd639b004340cd07"}},{"code_sha256_prefix":"e56544092cf931bd","entry":"binomial_sampler","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/common/sampler.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/common/sampler.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e56544092cf931bd"}},{"code_sha256_prefix":"5eec453ad4817229","entry":"convert_to_container","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/distribution/_base.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/distribution/_base.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5eec453ad4817229"}},{"code_sha256_prefix":"654eb783e0d83910","entry":"convert_to_module","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/distribution/_base.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/distribution/_base.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"654eb783e0d83910"}},{"code_sha256_prefix":"bcafdb774fca0219","entry":"get_path_var","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/common/env.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/common/env.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"bcafdb774fca0219"}},{"code_sha256_prefix":"3918737cb9502b15","entry":"mask_fill","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/common/torch_util.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/common/torch_util.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3918737cb9502b15"}},{"code_sha256_prefix":"9411bf15db0f1fa2","entry":"packed_attention_mask","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/common/torch_util.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/common/torch_util.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9411bf15db0f1fa2"}},{"code_sha256_prefix":"28ebfb3011b4bfaf","entry":"register_resolver","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/common/hydra_util.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/common/hydra_util.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"28ebfb3011b4bfaf"}},{"code_sha256_prefix":"568a5b54144b7946","entry":"resolve_as_tuple","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/common/hydra_util.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/common/hydra_util.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"568a5b54144b7946"}},{"code_sha256_prefix":"603ae909b9308e4e","entry":"resolve_cls_getattr","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/common/hydra_util.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/common/hydra_util.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"603ae909b9308e4e"}},{"code_sha256_prefix":"5ae38533a6777347","entry":"safe_div","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/common/torch_util.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/common/torch_util.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"5ae38533a6777347"}},{"code_sha256_prefix":"fa3ea93fdf7ea89a","entry":"tree_map_multi","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/distribution/_base.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/distribution/_base.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fa3ea93fdf7ea89a"}},{"code_sha256_prefix":"47a7fdcc8c3ab2d0","entry":"uniform_sampler","repo":"zqiao11/MSFT","repo_kind":"found_in_text","path":"src/uni2ts/common/sampler.py","file_url":"https://github.com/zqiao11/MSFT/blob/HEAD/src/uni2ts/common/sampler.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"47a7fdcc8c3ab2d0"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}