{"paper_id":"2505.05784","title":"FlowHFT: Imitation Learning via Flow Matching Policy for Optimal High-Frequency Trading under Diverse Market Conditions","authors":["Yang Li","Zhi Chen","Steve Yang"],"paper_date":"2025-05-09","paper_type":"perspective-policy","paper_url":"https://arxiv.org/pdf/2505.05784v3","page_url":"https://thequant.space/flowcharts/flowhft-imitation-learning-via-flow-matching-policy-for-opt/","scores":{"empirical_rigor":4,"math_complexity":8.5,"quadrant":"Lab Rats"},"proposes":{"status":"available","statement":"The paper proposes FlowHFT, a novel imitation learning framework based on flow matching policy for high‑frequency trading.","quote":"we propose the FlowHFT, a novel imitation learning framework based on flow matching policy.","summary":"High-frequency trading (HFT) is an investing strategy that continuously monitors market states and places bid and ask orders at millisecond speeds. Traditional HFT approaches fit models with historical data and assume that future market states follow similar patterns. This limits the effectiveness…","strategy":null},"assumptions":{"status":"available","items":[{"assumption":"Traditional HFT models assume specific stochastic processes for asset prices, stable order flow, and no sudden volatility spikes.","type":"model","stated":"explicit","quote":"such models achieve optimal performance only under restrictive assumptions, including specific stochastic processes for asset prices, stable order flows, and the absence of sudden volatility spikes."},{"assumption":"The state satisfies the Markov property.","type":"statistical","stated":"explicit","quote":"We generally assume the state St satisfies the Markov property."}]},"datasets":{"status":"available","items":[{"name":"Synthetic state‑action pairs for training the diffusion policy","provider":null,"kind":"synthetic","access":"unknown","period":null,"universe":null,"quote":"In total, we have gathered 3.24 million state-action pairs. These generated state-action pairs serve as the training dataset for the diffusion policy."}],"known":[],"sample":{"start":null,"end":null,"market":null,"asset_class":null,"frequency":null},"data_resolution":"synthetic"},"transaction_costs":{"status":"not_extracted","modeled":"unclear","headline_net_of_costs":null,"detail":null,"quote":null,"source":"none","rule":"R0","candidates":{"front_matter":"unclear","evidence_card":"unclear","evidence_card_quoted":false},"meaning":"the classifier was not confident"},"results":{"status":"available","evaluation_type":"trading-pnl","findings":[{"finding":"FlowHFT consistently outperforms the best‑performing expert strategy in each tested market scenario.","kind":"performance","quote":"our unified framework consistently outperforms the best-performing expert strategy in each tested market scenario."},{"finding":"FlowHFT exhibits significant robustness to market conditions involving sudden price jumps.","kind":"performance","quote":"The results indicate that our model exhibits significant robustness to market conditions involving sudden price jumps."},{"finding":"Increasing the number of expert instructors improves Sharpe Ratio and reduces Maximum Drawdown.","kind":"performance","quote":"This progressive improvement indicates that FlowHFT effectively learns and integrates new, valuable strategic information from additional experts."},{"finding":"FlowHFT can generate trading actions at millisecond speeds suitable for HFT tasks.","kind":"performance","quote":"Additionally, FlowHFT is designed for rapid inference, enabling the generation of trading actions at millisecond speeds for HFT tasks."}],"reported_performance":null,"out_of_sample":{"method":"unclear","headline_out_of_sample":null}},"independent_evidence":{"status":"none_found","supports":[],"contradicts":[],"qualifies":[],"compared_with":31,"citations":{"cited_by_in_corpus":4,"by_independent_authors":0,"how_cited":{"used_as_baseline":0,"uses_method_or_data":0,"extends":0,"discusses_findings":0,"mentions":0,"unclassified":4},"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":"2026-11-14T17:38:46Z"},"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/"}