Newest papers

SOTA: Stock Options Trading Agents Guided by Option-Implied Return Distributions

Yizhen Xie, Mengyang Liu · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2610.10407 · cited by 0 works (OpenAlex) · open-access PDF

French

Avec la croissance des marchés d'options et l'avancée de l'IA, les systèmes d'agents pour le trading d'options gagnent de plus en plus d'attention. Les agents basés sur le modèle linguistique peuvent réfléchir sur des informations contextuelles telles que les nouvelles, mais le trading d'options présente un problème de décision particulièrement difficile: une seule action peut avoir des milliers de contrats, et l'agent doit décider à la fois quels contrats doivent être négociés et comment les combiner. Les approches existantes réduisent souvent cette complexité en restreignant la politique à une structure de stratégie fixe, telle qu'un straddle, limitant leur capacité à changer de stratégie lorsque les conditions du marché changent. Nous présentons SOTA (Stock Options Trading Agents

Machine translation (Meta M2M100); the English original follows.

Abstract

As option markets grow and AI advances, agentic systems for option trading are gaining increasing attention. Language-model-based agents can reason over contextual information such as news, but option trading presents a particularly challenging decision problem: a single stock can have thousands of contracts, and the agent must decide both which contracts to trade and how to combine them. Existing approaches often sidestep this complexity by restricting the policy to a fixed strategy structure, such as a straddle, limiting their ability to switch strategies as market conditions change. We present SOTA (Stock Options Trading Agents), an agentic trading framework for structured option-strategy selection. SOTA abstracts the large option universe into strategy-level decisions while deterministic resolvers handle portfolio implementation. We develop SOTA by post-training Qwen3.8-27B with supervised fine-tuning followed by reinforcement learning. SOTA is evaluated on options on nine large-cap U.S. equities and SPY against rule-based and machine-learning strategy selectors in the same trading environment. Over a six-month out-of-sample period, SOTA earns an 18.3% total return with a Sharpe ratio of 1.60 and a maximum drawdown of 8.96%. We also document an asymmetric role of news: news improves frontier-teacher trajectories, but retaining news during reinforcement learning reduces out-of-sample return from 18.3% to -2.7%.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Reinforcement learning, Portfolio, Trading strategy, Sharpe ratio, Economics, Limiting

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange