{"paper_id":"2509.11420","title":"Trading-R1: Financial Trading with LLM Reasoning via Reinforcement Learning","authors":["Yijia Xiao","Edward Sun","Tong Chen","Fang Wu","Di Luo","Wei Wang"],"paper_date":"2025-09-14","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2509.11420v1","page_url":"https://thequant.space/flowcharts/trading-r1-financial-trading-with-llm-reasoning-via-reinfor/","scores":{"empirical_rigor":6,"math_complexity":4.5,"quadrant":"Street Traders"},"proposes":{"status":"available","statement":"The paper proposes Trading-R1, a financially-aware LLM that integrates strategic reasoning, thesis composition, facts‑grounded analysis, and volatility‑adjusted decision making, trained via supervised fine‑tuning and reinforcement learning.","quote":"We present Trading-R1, a financially-aware model that incorporates strategic thinking and planning for comprehensive thesis composition, facts-grounded analysis, and volatility-adjusted decision making.","summary":"Developing professional, structured reasoning on par with human financial analysts and traders remains a central challenge in AI for finance, where markets demand interpretability and trust. Traditional time-series models lack explainability, while LLMs face challenges in turning natural-language…","strategy":"Trading-R1"},"assumptions":{"status":"available","items":[{"assumption":"Historical price and volume data are taken from Yahoo Finance with a 15‑day rolling window per trading day.","type":"data","stated":"implied","quote":"We obtained historical price and volume data from the Yahoo Finance API. For each trading day, a 15-day rolling window was extracted containing open, high, low, close (OHLC) values and trading volume."},{"assumption":"Trading decisions are evaluated with a neutral band and trading cost term κ in the outcome reward.","type":"execution","stated":"implied","quote":"directionally correct calls earn a magnitude- and intensity-calibrated fraction of b and pay trading cost κ"}]},"datasets":{"status":"available","items":[{"name":"Tauric-TR1-DB","provider":null,"kind":"text-news","access":"public","period":"Jan 1 2024 – May 31 2025","universe":"14 major equities","quote":"Training uses Tauric-TR1-DB, a 100k-sample corpus spanning 18 months, 14 equities, and five heterogeneous financial data sources."},{"name":"Yahoo Finance price data","provider":"Yahoo Finance","kind":"market-prices","access":"public","period":null,"universe":null,"quote":"We obtained historical price and volume data from the Yahoo Finance API.","known_id":"yahoo-finance"},{"name":"Technical indicators (stockstats)","provider":"stockstats library","kind":"alternative","access":"public","period":null,"universe":null,"quote":"In addition to raw prices, we computed a suite of widely used technical indicators using the stockstats library."}],"known":[{"id":"yahoo-finance","name":"Yahoo Finance","provider":"Yahoo","kind":"market-prices","access":"public","quote":"Price Data (Yahoo Finance) We obtained historical price and volume data from the Yahoo Finance API.","mentions":5,"used_in":1},{"id":"fred","name":"FRED","provider":"Federal Reserve Bank of St. Louis","kind":"macro","access":"public","quote":"FRED API Collection We sourced U.S. macro indicators from the Federal Reserve Economic Data (FRED) API using an authenticated key.","mentions":3,"used_in":1}],"sample":{"start":"2024-06","end":"2024-08","market":"six major equities and ETFs","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"unclear","headline_net_of_costs":false,"detail":null,"quote":null,"source":"conflict","rule":"R4b","candidates":{"front_matter":"no","evidence_card":"partial","evidence_card_quoted":true},"conflict_kind":"graded","card_detail":"A trading cost term κ is included in the outcome reward to encourage calibrated selectivity.","card_quote":"directionally correct calls earn a magnitude- and intensity-calibrated fraction of b and pay trading cost κ","meaning":"the platform classifier and the paper-text extraction disagree, so no value is asserted"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"Trading-R1 achieves a Sharpe ratio of 1.88 with 8.08% returns on NVDA and outperforms GPT‑4.1 on AAPL with a Sharpe of 1.80 versus 1.24, while maintaining lower drawdowns (3.68% vs. 7.88%).","kind":"performance","quote":"It achieves a Sharpe ratio of 1.88 with 8.08% returns on NVDA, and outperforms GPT-4.1 on AAPL with a Sharpe of 1.80 versus 1.24, while maintaining lower drawdowns (3.68% vs. RLMs’ 7.88%)."},{"finding":"Trading-R1 attains leading hit ratios, including 70.0% on NVDA and 64.0% on SPY.","kind":"performance","quote":"The model also attains leading hit ratios, including 70.0% on NVDA and 64.0% on SPY."},{"finding":"Across all evaluated assets, Trading-R1 delivers improvements over baseline models, achieving the strongest overall performance among SLM, RLM, LLM, Trading‑SFT, and Trading‑RFT.","kind":"performance","quote":"Our Trading-R1 approach achieves the strongest overall performance by combining SFT and RFT to capture market dynamics effectively."}],"reported_performance":{"strategy":"Trading-R1","sharpe":1.88,"sharpe_benchmark":null,"annual_return_pct":8.08,"total_return_pct":null,"max_drawdown_pct":3.68},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":30,"citations":{"cited_by_in_corpus":6,"by_independent_authors":6,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":6,"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://github.com/TauricResearch/Trading-R1","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":"2026-11-14T17:27:02Z"},"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/"}