{"paper_id":"2508.02366","title":"Language Model Guided Reinforcement Learning in Quantitative Trading","authors":["Adam Darmanin","Vince Vella"],"paper_date":"2025-08-04","paper_type":"empirical","paper_url":"https://arxiv.org/pdf/2508.02366v3","page_url":"https://thequant.space/flowcharts/language-model-guided-reinforcement-learning-in-quantitative/","scores":{"empirical_rigor":6,"math_complexity":6.5,"quadrant":"Holy Grail"},"proposes":{"status":"available","statement":"The paper proposes a hybrid framework where large language models generate high‑level trading strategies to guide reinforcement‑learning agents.","quote":"This paper proposes a hybrid framework in which LLMs generate high-level trading strategies to guide RL agents.","summary":"Algorithmic trading requires short-term tactical decisions consistent with long-term financial objectives. Reinforcement Learning (RL) has been applied to such problems, but adoption is limited by myopic behaviour and opaque policies. Large Language Models (LLMs) offer complementary strategic…","strategy":"LLM+RL hybrid trading agent"},"assumptions":{"status":"available","items":[{"assumption":"Sharpe Ratio calculations assume a zero risk‑free rate.","type":"statistical","stated":"explicit","quote":"their SR assumes a zero risk-free rate."},{"assumption":"The hybrid design does not modify the underlying RL algorithm architecture.","type":"model","stated":"explicit","quote":"improving out‑of-sample performance without modifying the RL algorithm."}]},"datasets":{"status":"available","items":[{"name":"Multi‑modal trading dataset","provider":null,"kind":"market-prices","access":"unknown","period":"2012–2020","universe":null,"quote":"To support strategy generation, the LLM agents consumed a multi-modal dataset spanning 2012–2020, aligning with the benchmark’s dataset dates in [3]. The dataset includes traditional Open, High, Low, Close, and Volume (OHLCV) price data..."},{"name":"Interactive Brokers market data","provider":"Interactive Brokers","kind":"market-prices","access":"commercial","period":"2012–2020","universe":null,"quote":"Market data was sourced from Interactive Brokers1 and iVolatility2, including OHLCV time series, SPX and NDX index returns, the VIX index, and Options implied volatility (IV)."},{"name":"Fundamental data via SEC‑API and FRED","provider":"SEC‑API / FRED","kind":"fundamentals","access":"public","period":"2012–2020","universe":null,"quote":"Fundamental data, comprising firm-level financial ratios and macroeconomic indicators (e.g., GDP, PMI, interest rates), were retrieved via SEC-API3 and the FRED API4.","known_id":"fred"},{"name":"Technical analytics features","provider":"TA‑Lib","kind":"alternative","access":"public","period":"2012–2020","universe":null,"quote":"Analytics features were computed using TA-Lib5, applying rolling windows to extract indicators."},{"name":"News headlines","provider":"Alpaca","kind":"text-news","access":"commercial","period":"2012–2020","universe":null,"quote":"Alternative data consisting of news headlines were collected from Alpaca6 and processed into explanatory factors using few-shot LLM prompting..."}],"known":[{"id":"fred","name":"FRED","provider":"Federal Reserve Bank of St. Louis","kind":"macro","access":"public","quote":"Fundamental data, comprising firm-level financial ratios and macroeconomic indicators (e.g., GDP, PMI, interest rates), were retrieved via SEC-API3 and the FRED API4.","mentions":1,"used_in":1}],"sample":{"start":"2018","end":"2020","market":"US large‑cap stocks","asset_class":"equities","frequency":"daily"},"data_resolution":"daily"},"transaction_costs":{"status":"available","modeled":"no","headline_net_of_costs":false,"detail":null,"quote":null,"source":"front_matter","rule":"R1","candidates":{"front_matter":"no","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"a trading or execution result is reported without costs"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"LLM guidance improves both return and risk metrics relative to standard RL.","kind":"performance","quote":"Empirical results indicate that LLM guidance improves both return and risk metrics relative to standard RL."},{"finding":"The hybrid LLM+RL agent out‑performed the RL‑only baseline in four out of six assets with statistically significant improvements (p < 0.05).","kind":"performance","quote":"All resulting p-values were below 0.05, indicating statistically significant improvements."}],"reported_performance":{"strategy":"LLM+RL hybrid trading agent","sharpe":1.7,"sharpe_benchmark":0.64,"annual_return_pct":null,"total_return_pct":null,"max_drawdown_pct":0.31},"out_of_sample":{"method":"holdout","headline_out_of_sample":true}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":15,"citations":{"cited_by_in_corpus":2,"by_independent_authors":2,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":2,"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":"2026-11-14T17:30:42Z"},"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/"}