Paper: arXiv 2609.33470

Authors: Haochen Luo, Yifan Li, Binh Minh An, Xiaolong Luo, Zhengzhao Lai, Yuan Zhang, Chen Liu

Abstract

Large language models (LLMs) and multi-agent systems (MAS) have shown promise in financial decision-making, yet existing evaluations focus on equity trading and primarily assess directional prediction, overlooking the structural complexity of derivative markets. Option trading introduces fundamentally different challenges, including nonlinear payoffs and multi-leg strategy construction, requiring structured decisions rather than simple directional bets. We introduce LiveOption, an evaluation framework for LLM-based agents in option trading. LiveOption formulates the problem as structured sequential decision-making under realistic execution and capital constraints, and provides a reproducible environment with standardized interaction protocols. The framework includes three task suites covering portfolio overlays, event-driven earnings trading, and 0DTE intraday trading. We further propose a hierarchical metric suite that evaluates action validity, decision quality, risk characteristics, and outcome-level performance. Experiments show that current agents often fail to achieve competitive returns in most scenarios. LiveOption offers a principled testbed for evaluating structured decision-making beyond outcome-based metrics.

Complexity vs Empirical Score

  • Math Complexity: 6.0/10
  • Empirical Rigor: 8.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: This paper introduces a novel and much-needed evaluation framework for LLM agents in the complex domain of option trading, moving beyond simple equity predictions. It demonstrates strong empirical rigor through its detailed backtesting engine, multi-scenario evaluation, and hierarchical metric suite. The framework’s design for reproducibility and its focus on structured decision-making under realistic constraints are significant strengths.

Research Flowchart

  flowchart TD
    A["Research Goal: Evaluate LLM Agents in Structured Option Trading"] --> B("Methodology: LiveOption Framework Development")
    B --> C{"Data/Inputs: Realistic Market Data, Execution & Capital Constraints"}
    C --> D["Computational Processes: LLM Agent Decision-Making, Multi-leg Strategy Construction, Nonlinear Payoff Calculation"]
    D --> E{"Evaluation: Hierarchical Metric Suite - Validity, Quality, Risk, Outcome"}
    E --> F["Key Finding: Current LLM Agents Struggle to Achieve Competitive Returns in Option Trading"]

Evidence Card

What the authors claim, on what data, and who has checked it. Read from the paper's own text by a language model: every quote is verbatim, nothing was re-run. How to read this card

Proposes

The paper proposes LiveOption, an evaluation framework for LLM-driven option trading agents.

Source sentence
We introduce LiveOption, an evaluation framework for LLM-based agents in option trading.

Assumes 3

  • Option markets provide a more realistic stress test requiring reasoning about volatility dynamics, tail risk, capital efficiency, and structured strategies. market
    Source sentence
    our central hypothesis is that option markets provide a more realistic and demanding stress test for financial agents than stock-only environments, as they require simultaneous reasoning about volatility dynamics, tail risk, capital efficiency, and the construction of structured strategies.
  • The backtesting engine simulates realistic brokerage-style execution with margin enforcement and multi‑leg position management. execution
    Source sentence
    it incorporates a fully-featured backtesting engine that simulates realistic brokerage-style execution, including margin enforcement and multi-leg position management
  • Trading costs are modeled as option commission $0.65 per contract, stock commission $0.0005 per share, slippage 1.0 bp, and mid‑quote fills. execution
    Source sentence
    Option commission $0.65 per contract Stock commission $0.0005 per share Slippage 1.0 bp applied to fill price Bid–ask spread Mid-quote fills (0 bp explicit spread cost)

Data

Backtest sample: 2025-01 – 2025-12 · US equities (QQQ, NVDA, AMD, GOOG, TSLA) · options · daily

  • Overlay underlyings options unknown
    2025 fiscal year · QQQ, NVDA, AMD, GOOG, TSLA
    Source sentence
    The five overlay underlyings, QQQ, NVDA, AMD, GOOG, and TSLA, are chosen to span the regimes a single-name option overlay must be evaluated against.

Costs

Modeled Costs are part of the reported headline results.

Option commission $0.65 per contract, stock commission $0.0005 per share, slippage 1.0 bp, mid‑quote fills.

Source sentence
Option commission $0.65 per contract Stock commission $0.0005 per share Slippage 1.0 bp applied to fill price Bid–ask spread Mid-quote fills (0 bp explicit spread cost)

Finds 1

Reported performance (GPT-OSS-120B option overlay): Sharpe 1.01

  • Current LLM agents achieve negative episode returns across all tasks, with no statistically significant cross‑model differences. negative
    Source sentence
    Our results reveal systematic limitations in current LLM agents: typical episode returns are negative in every task, no cross-model difference survives correction for multiple comparisons

Checked by others

No other paper in the archive has been found to support or contradict these findings (compared with 5 related papers). Coverage is partial, so this is neither replication nor refutation.

Code

No repository by the authors found in the paper.

Extracted 2026-11-14 · same card as JSON · A language model chose these items: check the source sentence before relying on one.

Cited References (89)

  1. Leveraging large language models for predicting stock option valuation and financial risk mitigation (2024)
  2. Deep reinforcement learning for active high frequency trading (2021) — link
  3. Beat the Market An Effective Intraday Momentum Strategy for S&P500 ETF (SPY) (2024)
  4. Does 0DTE Options Trading Increase Volatility? (2023)
  5. Trading Options at Expiration: Strategies and Models for Winning the Endgame (2009)
  6. Day Trading Options: Profiting from Price Distortions in Very Brief Time Frames (2009)
  7. Simulating Dynamics of the Recognition-Primed Decision Model on Financial Markets Using LLM (2026)
  8. Machine learning methods for pricing financial derivatives — → 2406.00459
and 81 more
  1. Developing AI-Powered Predictive Analytics for Derivatives Trading: Machine Learning Models for Market Volatility Forecasting, Risk Mitigation, and Strategy Optimization (2023)
  2. Implied volatility directional forecasting: a machine learning approach (2021)
  3. AI-Trader: Benchmarking Autonomous Agents in Real-Time Financial Markets — → 2512.10971
  4. Learning to Use Tools via Cooperative and Interactive Agents (2024) — link
  5. Multi-agent collaboration mechanisms: A survey of llms (2025) — link
  6. A survey of agent interoperability protocols: Model context protocol (mcp), agent communication protocol (acp), agent-to-agent protocol (a2a), and agent network protocol (anp) (2025) — link
  7. Docfinqa: A long-context financial reasoning dataset (2024)
  8. Pixiu: A comprehensive benchmark, instruction dataset and large language model for finance (2023)
  9. Finqa: A dataset of numerical reasoning over financial data (2021)
  10. On the estimation of security price volatilities from historical data (1980)
  11. AlphaBench: Benchmarking Large Language Models in Formulaic Alpha Factor Mining (2026)
  12. Tradinggpt: Multi-agent system with layered memory and distinct characters for enhanced financial trading performance — → 2309.03736
  13. Large Language Models as Tool Makers (2024) — link
  14. React: Synergizing reasoning and acting in language models (2022)
  15. Toolformer: Language Models Can Teach Themselves to Use Tools (2023) — link
  16. Gorilla: Large Language Model Connected with Massive API s (2024) — link
  17. Model context protocol (mcp): Landscape, security threats, and future research directions (2025)
  18. QF in Z ero: A Unified Financial Toolchain for LLM -Based Trading Agents (2026) — link
  19. A survey on model context protocol: Architecture, state-of-the-art, challenges and future directions (2025)
  20. TradingAgents: Multi-Agents LLM Financial Trading Framework — → 2412.20138
  21. A comprehensive survey of contamination detection methods in large language models (2024) — link
  22. Leak, cheat, repeat: Data contamination and evaluation malpractices in closed-source LLMs (2024)
  23. Constat: Performance-based contamination detection in large language models (2024)
  24. Finmem: A performance-enhanced llm trading agent with layered memory and character design — → 2311.13743
  25. Cfbenchmark: Chinese financial assistant benchmark for large language model (2023) — link
  26. Financebench: A new benchmark for financial question answering (2023) — link
  27. Finben: A holistic financial benchmark for large language models (2024)
  28. Investorbench: A benchmark for financial decision-making tasks with llm-based agent — → 2412.18174
  29. Flag-trader: Fusion llm-agent with gradient-based reinforcement learning for financial trading — → 2502.11433
  30. CryptoTrade: A reflective LLM-based agent to guide zero-shot cryptocurrency trading (2024)
  31. Qwen2.5 technical report (2024) — link
  32. Benchmarking benchmark leakage in large language models (2024) — link
  33. The memorization problem: Can we trust llms’ economic forecasts? — → 2504.14765
  34. Assessing look-ahead bias in stock return predictions generated by gpt sentiment analysis — → 2309.17322
  35. Evaluating LLMs in Finance Requires Explicit Bias Consideration — → 2602.14233
  36. SquidAgent: Parallelize Wisely, Coordinate Efficiently (2026) — link
  37. C laude S onnet 4 — anthropic.com
  38. GPT-5.6 Terra Model (2026)
  39. Gemini 3 Flash (2026)
  40. When Agents Trade: Live Multi-Market Trading Arena for LLM Agents (2026)
  41. Stockbench: Can llm agents trade stocks profitably in real-world markets? (2025) — link
  42. Gpt-4 technical report (2023) — link
  43. Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities (2025) — link
  44. Deepseek-v3 technical report (2024) — link
  45. Forecasting implied volatility smile surface via deep learning and attention mechanism (2019) — link
  46. Improving option price forecasts with neural networks and support vector regressions (2009)
  47. Financial density forecasts: A comprehensive comparison of risk-neutral and historical schemes (2018)
  48. The pricing of options and corporate liabilities (1973)
  49. A neural network versus Black–Scholes: a comparison of pricing and hedging performances (2003)
  50. Deep hedging of derivatives using reinforcement learning (2021) — link
  51. Retail Traders Love 0DTE Options… But Should They? (2023)
  52. Pricing and hedging short sterling options using neural networks (2012)
  53. Deep learning for options trading: an end-to-end approach — → 2407.21791
  54. HedgeAgents: A Balanced-aware Multi-agent Financial Trading System — → 2502.13165
  55. Dynamic Hedging Strategies in Derivatives Markets with LLM-Driven Sentiment and News Analytics (2025)
  56. Options Pricing Platform with Neural Networks, LLMs and Reinforcement Learning (2025)
  57. Kronos: A Foundation Model for the Language of Financial Markets — → 2508.02739
  58. Stocktime: A time series specialized large language model architecture for stock price prediction — → 2409.08281
  59. Alpha-gpt: Human-ai interactive alpha mining for quantitative investment — → 2308.00016
  60. Stock Market Forecasting: From Traditional Predictive Models to Large Language Models: M. Darwish et al. (2025)
  61. Futurex: An advanced live benchmark for llm agents in future prediction (2025) — link
  62. FinVision: A multi-agent framework for stock market prediction — → 2411.08899
  63. Comparative analysis of LLM-based market prediction and human expertise with sentiment analysis and machine learning integration (2024)
  64. Cognitive Alpha Mining via LLM-Driven Code-Based Evolution (2025) — link
  65. AlphaAgent: LLM-driven alpha mining with regularized exploration to counteract alpha decay (2025)
  66. Navigating the alpha jungle: an LLM-Powered MCTS framework for formulaic factor mining (2025) — link
  67. Quantagent: Price-driven multi-agent llms for high-frequency trading (2025) — link
  68. From Natural Language to Executable Option Strategies via Large Language Models — → 2603.16434
  69. TwinMarket: A Scalable Behavioral and Social Simulation for Financial Markets (2025) — link
  70. Fincon: A synthesized llm multi-agent system with conceptual verbal reinforcement for enhanced financial decision making (2024)
  71. Fintextqa: A dataset for long-form financial question answering (2024)
  72. Kelly-based options trading strategies on settlement date via supervised learning algorithms (2022)
  73. OPHR : Mastering Volatility Trading with Multi-Agent Deep Reinforcement Learning (2025) — link
  74. Time Travel is Cheating: Going Live with DeepFund for Real-Time Fund Investment Benchmarking (2025) — link
  75. DeepSeek-V4 : Towards Highly Efficient Million-Token Context Intelligence (2026) — link
  76. gpt-oss-120b & gpt-oss-20b Model Card (2025) — link
  77. Qwen3 Technical Report (2025) — link
  78. The Llama 3 Herd of Models (2024) — link
  79. MiniMax Sparse Attention (2026) — link
  80. GLM-5 : from Vibe Coding to Agentic Engineering (2026) — link
  81. GLM-5.3 — Model Overview (2026)

17 of 89 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2596 linked papers.