{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,3]],"date-time":"2026-06-03T15:49:29Z","timestamp":1780501769980,"version":"3.54.1"},"reference-count":24,"publisher":"Association for Computing Machinery (ACM)","issue":"2","license":[{"start":{"date-parts":[[2025,1,21]],"date-time":"2025-01-21T00:00:00Z","timestamp":1737417600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["SIGKDD Explor. Newsl."],"published-print":{"date-parts":[[2025,1,21]]},"abstract":"<jats:p>Large language models (LLMs) have been applied in many fields and have developed rapidly in recent years. As a classic machine learning task, time series forecasting has recently been boosted by LLMs. Recent works treat large language models as zero-shot time series reasoners without further fine-tuning, which achieves remarkable performance. However, some unexplored research problems exist when applying LLMs for time series forecasting under the zero-shot setting. For instance, the LLMs' preferences for the input time series are less understood. In this paper, by comparing LLMs with traditional time series forecasting models, we observe many interesting properties of LLMs in the context of time series forecasting. First, our study shows that LLMs perform well in predicting time series with clear patterns and trends but face challenges with datasets lacking periodicity. This observation can be explained by the ability of LLMs to recognize the underlying period within datasets, which is supported by our experiments. In addition, the input strategy is investigated, and it is found that incorporating external knowledge and adopting natural language paraphrases substantially improves the predictive performance of LLMs for time series. Our study contributes insight into LLMs' advantages and limitations in time series forecasting under different conditions.<\/jats:p>","DOI":"10.1145\/3715073.3715083","type":"journal-article","created":{"date-parts":[[2025,1,29]],"date-time":"2025-01-29T23:19:36Z","timestamp":1738192776000},"page":"109-118","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":31,"title":["Time Series Forecasting with LLMs: Understanding and Enhancing Model Capabilities"],"prefix":"10.1145","volume":"26","author":[{"given":"Hua","family":"Tang","sequence":"first","affiliation":[{"name":"Shanghai Jiaotong University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chong","family":"Zhang","sequence":"additional","affiliation":[{"name":"University of Liverpool"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mingyu","family":"Jin","sequence":"additional","affiliation":[{"name":"Rutgers University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qinkai","family":"Yu","sequence":"additional","affiliation":[{"name":"University of Liverpool"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhenting","family":"Wang","sequence":"additional","affiliation":[{"name":"Rutgers University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaobo","family":"Jin","sequence":"additional","affiliation":[{"name":"Xi?an Jiaotong-Liverpool University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongfeng","family":"Zhang","sequence":"additional","affiliation":[{"name":"Rutgers University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Mengnan","family":"Du","sequence":"additional","affiliation":[{"name":"New Jersey Institute of Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2025,1,29]]},"reference":[{"key":"e_1_2_1_1_1","volume-title":"Magoc et al., \"A study of generative large language model for medical research and healthcare,\" NPJ Digital Medicine","author":"Peng C.","year":"2023","unstructured":"C. Peng, X. Yang, A. Chen, K. E. Smith, N. Pour Nejatian, A. B. Costa, C. Martin, M. G. Flores, Y. Zhang, T. Magoc et al., \"A study of generative large language model for medical research and healthcare,\" NPJ Digital Medicine, 2023."},{"key":"e_1_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.1108\/JBIM-07-2021-0332"},{"key":"e_1_2_1_3_1","volume-title":"FinBERT: A large language model for extracting information from financial text,\" Contemporary Accounting Research","author":"Huang A. H.","year":"2023","unstructured":"A. H. Huang, H. Wang, and Y. Yang, \"FinBERT: A large language model for extracting information from financial text,\" Contemporary Accounting Research, 2023."},{"key":"e_1_2_1_4_1","volume-title":"Du et al., \"The impact of reasoning step length on large language models,\" arXiv preprint arXiv:2401.04925","author":"Jin M.","year":"2024","unstructured":"M. Jin, Q. Yu, H. Zhao, W. Hua, Y. Meng, Y. Zhang, M. Du et al., \"The impact of reasoning step length on large language models,\" arXiv preprint arXiv:2401.04925, 2024."},{"key":"e_1_2_1_5_1","volume-title":"Large language models are zero-shot time series forecasters,\" arXiv preprint arXiv:2310.07820","author":"Gruver N.","year":"2023","unstructured":"N. Gruver, M. Finzi, S. Qiu, and A. G. Wilson, \"Large language models are zero-shot time series forecasters,\" arXiv preprint arXiv:2310.07820, 2023."},{"key":"e_1_2_1_6_1","volume-title":"Schneider et al., \"Lag-LLAMA: Towards foundation models for time series forecasting,\" arXiv preprint arXiv:2310.08278","author":"Rasul K.","year":"2023","unstructured":"K. Rasul, A. Ashok, A. R. Williams, A. Khorasani, G. Adamopoulos, R. Bhagwatkar, M. Bilo?, H. Ghon\u00eda, N. V. Hassen, A. Schneider et al., \"Lag-LLAMA: Towards foundation models for time series forecasting,\" arXiv preprint arXiv:2310.08278, 2023."},{"key":"e_1_2_1_7_1","volume-title":"TEST: Text-prototype aligned embedding to activate LLM?s ability for time series,\" arXiv preprint arXiv:2308.08241","author":"Sun C.","year":"2023","unstructured":"C. Sun, Y. Li, H. Li, and S. Hong, \"TEST: Text-prototype aligned embedding to activate LLM?s ability for time series,\" arXiv preprint arXiv:2308.08241, 2023."},{"key":"e_1_2_1_8_1","unstructured":"HuggingFace \"Chapter 6.5 of NLP course \" 2023 accessed: 2023-02--10."},{"key":"e_1_2_1_9_1","article-title":"STL: A seasonal-trend decomposition","author":"Cleveland R. B.","year":"1990","unstructured":"R. B. Cleveland, W. S. Cleveland, J. E. McRae, and I. Terpenning, \"STL: A seasonal-trend decomposition,\" J. Off. Stat, 1990.","journal-title":"J. Off. Stat"},{"issue":"124","key":"e_1_2_1_10_1","first-page":"1","article-title":"DARTS: User-friendly modern machine learning for time series","volume":"23","author":"Herzen J.","year":"2022","unstructured":"J. Herzen, F. L\u00e4ssig, S. G. Piazzetta, T. Neuer, L. Tafti, G. Raille, T. Van Pottelbergh, M. Pasieka, A. Skrodzki, N. Huguenin et al., \"DARTS: User-friendly modern machine learning for time series,\" Journal of Machine Learning Research, vol. 23, no. 124, pp. 1--6, 2022.","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_1_11_1","volume-title":"Monash time series forecasting archive,\" NeurIPS Systems Track on Datasets and Benchmarks","author":"Godahewa R. W.","year":"2021","unstructured":"R. W. Godahewa, C. Bergmeir, G. Webb, R. Hyndman, and P. Montero-Manso, \"Monash time series forecasting archive,\" NeurIPS Systems Track on Datasets and Benchmarks, 2021."},{"key":"e_1_2_1_12_1","first-page":"2021","article-title":"Informer: Beyond efficient transformer for long sequence time-series forecasting","author":"Zhou H.","year":"2021","unstructured":"H. Zhou, S. Zhang, J. Peng, S. Zhang, J. Li, H. Xiong, and W. Zhang, \"Informer: Beyond efficient transformer for long sequence time-series forecasting,\" AAAI 2021, 2021.","journal-title":"AAAI"},{"key":"e_1_2_1_13_1","volume-title":"Characteristic-based clustering for time series data,\" Data Mining and Knowledge Discovery","author":"Wang X.","year":"2006","unstructured":"X. Wang, K. Smith, and R. Hyndman, \"Characteristic-based clustering for time series data,\" Data Mining and Knowledge Discovery, 2006."},{"key":"e_1_2_1_14_1","doi-asserted-by":"crossref","unstructured":"M. S. Bartlett \"Periodogram analysis and continuous spectra \" Biometrika 1950.","DOI":"10.2307\/2332141"},{"key":"e_1_2_1_15_1","first-page":"2023","article-title":"Numeric magnitude comparison effects in large language models","author":"Shah R.","year":"2023","unstructured":"R. Shah, V. Marupudi, R. Koenen, K. Bhardwaj, and S. Varma, \"Numeric magnitude comparison effects in large language models,\" Findings of ACL 2023, 2023.","journal-title":"Findings of ACL"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1080\/01621459.1970.10481180"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ijforecast.2006.03.005"},{"key":"e_1_2_1_18_1","volume-title":"TimeGPT-1,\" arXiv preprint arXiv:2310.03589","author":"Garza A.","year":"2023","unstructured":"A. Garza and M. Mergenthaler-Canseco, \"TimeGPT-1,\" arXiv preprint arXiv:2310.03589, 2023."},{"key":"e_1_2_1_19_1","volume-title":"A decoder-only foundation model for time-series forecasting,\" arXiv preprint arXiv:2310.10688","author":"Das A.","year":"2023","unstructured":"A. Das, W. Kong, R. Sen, and Y. Zhou, \"A decoder-only foundation model for time-series forecasting,\" arXiv preprint arXiv:2310.10688, 2023."},{"key":"e_1_2_1_20_1","volume-title":"Tempo: Prompt-based generative pre-trained transformer for time series forecasting,\" arXiv preprint arXiv:2310.04948","author":"Cao D.","year":"2023","unstructured":"D. Cao, F. Jia, S. O. Arik, T. Pfister, Y. Zheng, W. Ye, and Y. Liu, \"Tempo: Prompt-based generative pre-trained transformer for time series forecasting,\" arXiv preprint arXiv:2310.04948, 2023."},{"key":"e_1_2_1_21_1","volume-title":"Aligning pre-trained LLMs as data-efficient time-series forecasters","author":"Chang C.","year":"2024","unstructured":"C. Chang, W.-Y. Wang, W.-C. Peng, and T.-F. Chen, \"LLM4TS: Aligning pre-trained LLMs as data-efficient time-series forecasters,\" 2024."},{"key":"e_1_2_1_22_1","volume-title":"Goal-guided generative prompt injection attack on large language models,\" arXiv preprint arXiv:2404.07234","author":"Zhang C.","year":"2024","unstructured":"C. Zhang, M. Jin, Q. Yu, C. Liu, H. Xue, and X. Jin, \"Goal-guided generative prompt injection attack on large language models,\" arXiv preprint arXiv:2404.07234, 2024."},{"key":"e_1_2_1_23_1","volume-title":"Pan et al., \"Time-LLM: Time series forecasting by reprogramming large language models,\" arXiv preprint arXiv:2310.01728","author":"Jin M.","year":"2023","unstructured":"M. Jin, S. Wang, L. Ma, Z. Chu, J. Y. Zhang, X. Shi, P.-Y. Chen, Y. Liang, Y.-F. Li, S. Pan et al., \"Time-LLM: Time series forecasting by reprogramming large language models,\" arXiv preprint arXiv:2310.01728, 2023."},{"key":"e_1_2_1_24_1","article-title":"PromptCast: A new prompt-based learning paradigm for time series forecasting","author":"Xue H.","year":"2023","unstructured":"H. Xue and F. D. Salim, \"PromptCast: A new prompt-based learning paradigm for time series forecasting,\" IEEE Transactions on Knowledge and Data Engineering, 2023.","journal-title":"IEEE Transactions on Knowledge and Data Engineering"}],"container-title":["ACM SIGKDD Explorations Newsletter"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3715073.3715083","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3715073.3715083","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T01:56:53Z","timestamp":1750298213000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3715073.3715083"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,1,21]]},"references-count":24,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2025,1,21]]}},"alternative-id":["10.1145\/3715073.3715083"],"URL":"https:\/\/doi.org\/10.1145\/3715073.3715083","relation":{},"ISSN":["1931-0145","1931-0153"],"issn-type":[{"value":"1931-0145","type":"print"},{"value":"1931-0153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,1,21]]},"assertion":[{"value":"2025-01-29","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}