{"paper_id":"2409.19854","title":"The Construction of Instruction-tuned LLMs for Finance without Instruction Data Using Continual Pretraining and Model Merging","authors":["Masanori Hirano","Kentaro Imajo"],"paper_date":"2024-09-30","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2409.19854v1","page_url":"https://thequant.space/flowcharts/the-construction-of-instruction-tuned-llms-for-finance-witho/","scores":{"empirical_rigor":6,"math_complexity":2,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes a method that builds instruction‑tuned financial LLMs by first doing domain‑specific continual pretraining and then merging the resulting model with a publicly available instruction‑tuned LLM.","quote":"Our study proposes a simpler approach that combines domain-specific continual pretraining with model merging.","summary":"This paper proposes a novel method for constructing instruction-tuned large language models (LLMs) for finance without instruction data. Traditionally, developing such domain-specific LLMs has been resource-intensive, requiring a large dataset and significant computational power for continual…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Instruction support and domain‑specific knowledge are independent in task arithmetic.","type":"model","stated":"explicit","quote":"For the success of the model merging, we assumed that instruction support and domain-specific knowledge are independent in task arithmetic."}]},"datasets":{"status":"available","items":[{"name":"Japanese financial corpus","provider":null,"kind":"text-news","access":"public","period":null,"universe":null,"quote":"The crawled articles mainly included the following types of documents: • speeches and press conferences of officers of the Bank of Japan • minutes of the monetary policy meetings of the Bank of Japan • reports, glossaries, and company profiles from multiple financial institutions • financial documents extracted from Wi"}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"none"},"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":"The domain‑specific instruction‑tuned LLM achieved the best performance on benchmarks, though the gap to general‑purpose instruction‑tuned LLMs was small.","kind":"performance","quote":"the best performance. However, the difference between the domain-specific and general-purpose instruction-tuned LLMs is not great."},{"finding":"The proposed continual pretraining and model‑merging pipeline successfully constructs an instruction‑tuned LLM for finance without any instruction data.","kind":"performance","quote":"In this study, we successfully constructed the instruction-tuned LLM for finance without instruction data using continual pretraining and model merging."},{"finding":"Weight similarity analysis shows low cosine similarity between domain‑specific and instruction task‑arithmetic vectors, supporting the independence assumption.","kind":"theoretical","quote":"Fig. 4. Weight Similarities between the Domain-specific and Instruction Task Arithmetic Vectors of the Merged Model"}],"reported_performance":null,"out_of_sample":{"method":"none","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":11,"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":true,"code_url":"https://huggingface.co/pfnet/nekomata-14b-pfn-qfin-inst-merge","note":null},"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":"2026-11-14T18:00:49Z"},"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/"}