{"paper_id":"2506.06288","title":"DELPHYNE: A Pre-Trained Model for General and Financial Time Series","authors":["Xueying Ding","Aakriti Mittal","Achintya Gopal"],"paper_date":"2025-05-12","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2506.06288v1","page_url":"https://thequant.space/flowcharts/delphyne-a-pre-trained-model-for-general-and-financial-time/","scores":{"empirical_rigor":5.5,"math_complexity":7.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes Delphyne, a pre‑trained model for general and finance‑specific time‑series tasks.","quote":"we introduce a Pre-trained MoDEL for FINance TimE-series (Delphyne).","summary":"Time-series data is a vital modality within data science communities. This is particularly valuable in financial applications, where it helps in detecting patterns, understanding market behavior, and making informed decisions based on historical data. Recent advances in language modeling have led…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Probability forecasting assumes a fixed output distribution.","type":"model","stated":"explicit","quote":"For probability forecasting, previous studies usually assume a fixed output distribution (Salinas et al., 2020a)."},{"assumption":"A single distribution may be insufficient when data presents varying distributions and supports.","type":"model","stated":"explicit","quote":"However, when the data presents varying distributions and supports, relying on a single distribution may be insufficient."},{"assumption":"For simplicity, the Student’s T mixture is chosen as the output distribution.","type":"model","stated":"explicit","quote":"For simplicity (Occam’s Razor), we choose the Student’s T mixture as the output distribution, with further discussion in the probability quantification experiment (Sec. 5.4)."},{"assumption":"Zero‑shot evaluation uses a fixed context length of 500.","type":"execution","stated":"explicit","quote":"For zero-shot, we fix context length equals 500 and report the test performance."},{"assumption":"Fine‑tuning searches context length in {100, 300, 500, 1000, 2000} and learning rate in {5e‑5, 1e‑5}.","type":"execution","stated":"explicit","quote":"For fine-tuning, we search context length in {100, 300, 500, 1000, 2000} and the learning rate in {5e −5, 1e −5}."},{"assumption":"Synthetic pre‑training data are generated as Wavelet functions and GARCH‑style data.","type":"data","stated":"explicit","quote":"Pre-training with GARCH and Wavelet Data To simulate datasets encountered in real‑world scenarios, we generate two types of synthetic data: Wavelet functions and GARCH‑style data."}]},"datasets":{"status":"available","items":[{"name":"LOTSA","provider":null,"kind":"benchmark","access":"public","period":null,"universe":null,"quote":"Delphyne is trained on both LOTSA (the largest public times-series dataset (Woo et al., 2024)) and financial data (Sec.4)."},{"name":"Monash Time Series Forecasting Repository","provider":null,"kind":"benchmark","access":"public","period":null,"universe":null,"quote":"Monash The Monash Time Series Forecasting Repository (Godahewa et al., 2021) is a comprehensive collection of diverse time series datasets."},{"name":"Synthetic Wavelet functions","provider":null,"kind":"synthetic","access":"public","period":null,"universe":null,"quote":"functions are composed of a combination of sine and cosine waves."},{"name":"Synthetic GARCH‑style data","provider":null,"kind":"synthetic","access":"public","period":null,"universe":null,"quote":"in GARCH models the current time‑steps are based on past squared residuals and past"},{"name":"Financial returns of stocks across various sectors","provider":null,"kind":"market-prices","access":"public","period":null,"universe":null,"quote":"that Delphyne is trained on both strongly and weakly correlated financial data such as the returns of stocks across various sectors, it employs an any‑variate attention mechanism."},{"name":"Various public datasets included in LOTSA (e.g., BuildingsBench, ClimateLearn, CloudOps, G","provider":null,"kind":"benchmark","access":"public","period":null,"universe":null,"quote":"We include both publicly available data and financial data in Del‑phyne’s training, allowing Del‑phyne to gene…"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"intraday-bars"},"transaction_costs":{"status":"not_applicable","modeled":"n/a","headline_net_of_costs":null,"detail":null,"quote":null,"source":"front_matter","rule":"R3","candidates":{"front_matter":"n/a","evidence_card":"n/a","evidence_card_quoted":false},"meaning":"the paper reports no trading or execution result"},"results":{"status":"available","evaluation_type":"forecast-accuracy","findings":[{"finding":"Delphyne‑F slightly outperforms the Naive model in zero‑shot performance, while Delphyne‑A and Delphyne‑L achieve similar performance after fine‑tuning, rivaling the best model MOIRAI.","kind":"performance","quote":"In zero-shot performance, Delphyne-F, trained solely on finance data, slightly outperforms the Naive model and the other three benchmarks"},{"finding":"Negative transfer is observed when pre‑training includes data from disparate domains, reducing zero‑shot forecast quality.","kind":"negative","quote":"negative transfer can occur when cross‑domain data is added during pre‑training... The appearances of data from too different distributions can lead to less effective zero‑shot forecast results."},{"finding":"A mixture of Student’s T distributions performs similarly to the more complex MOIRAI mixture and better than a single distribution for stock return NLL.","kind":"predictive-accuracy","quote":"Fig. 4 shows that the Student’s T mixture performs similarly to MOIRAI’s Mixture on measuring stock returns. On the other hand, a single distribution underperforms both."},{"finding":"Architectural modifications, including any‑variate attention, improve handling of multivariate financial data without forced mixing.","kind":"performance","quote":"that Delphyne is trained on both strongly and weakly correlated financial data such as the returns of stocks across various sectors, it employs an any‑variate attention mechanism. This mechanism enables the integration of cross‑channel information without forcibly mixing variate data."},{"finding":"Delphyne consistently outperforms or matches state‑of‑the‑art baselines on both general and finance time‑series tasks.","kind":"performance","quote":"We conduct experiments for both general and finance time‑series tasks, demonstrating that Delphyne consistently outperforms or matches state‑of‑the‑art baselines."}],"reported_performance":null,"out_of_sample":{"method":"holdout","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":41,"citations":{"cited_by_in_corpus":1,"by_independent_authors":1,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":1,"unclassified":0},"independent_papers":[],"scope":"Citations between papers in this corpus only; citations from outside it are not counted."}},"implementation":{"status":"available","public":false,"code_url":null,"note":"No repository presented by the authors as their own was found in the paper. Absence here is not proof that no code exists."},"replication":{"status":"none_found","card":null,"note":"No replication of this paper has been run by thequant.space yet."},"extraction":{"text_source":"fulltext","model":"openai/gpt-oss-120b","version":1,"extracted_at":"2026-11-14T17:38:54Z"},"provenance":{"claims":"Everything here is what the authors state in the paper. Nothing was re-run or reproduced by the platform.","quotes":"Assumptions, datasets, cost detail and findings were extracted by a language model. Every quote is verbatim paper text. Where an item has `support`, \"full\" means every figure in it appears in the quoted sentences and \"partial\" means a figure is stated elsewhere in the paper; items with a figure found nowhere in the paper were discarded.","datasets_known":"datasets.known lists well-known vendors and public datasets by a stable id, found by a dictionary over the full text where the sentence says the data is used; it is independent of the model and covers only names in the dictionary.","flags":"transaction_costs.modeled, results.evaluation_type and results.out_of_sample.method are platform classifications (structured-decision models); see https://thequant.space/score-guide/#penalty-flags","comparisons":"supports / contradicts / qualifies are a language model's reading of two papers' quoted findings, restricted to closely related papers that share no author. They are pointers to read, not verified replications, and coverage is partial.","implementation":"code_url is a repository the paper itself presents as its own (rule-scored, ambiguous cases judged by a model), checked for liveness when it was recorded."},"docs":"https://thequant.space/agents/"}