Paper: arXiv 2604.26747

Authors: Yikuan Huang, Zheqi Fan, Kaiqi Hu, Yifan Ye

Abstract

LLM agents are promising tools for empirical discovery, but their flexibility can also turn discovery into uncontrolled search. We study how to use agents under a reproducible protocol through cryptocurrency factor discovery. Our framework casts the task as sequential hypothesis search: an agent reads an append-only experiment trace, proposes falsifiable factor hypotheses, and maps them to executable recipes, while a deterministic engine enforces fixed data splits, selection gates, transaction costs, and portfolio tests. Candidate actions are restricted to a point-in-time factor DSL, making both successful and failed hypotheses auditable. A ridge-combined portfolio trained only on 2020–2022 data achieves a 44.55% annualized return and Sharpe ratio of 1.55 in the 2024–2026 pure out-of-sample period after a 5 basis point one-way trading cost.

Complexity vs Empirical Score

  • Math Complexity: 4.5/10
  • Empirical Rigor: 8.5/10
  • Quadrant: Street Traders — practical and empirical, lighter on theory

Why this score: The paper focuses on building a robust, auditable, and backtest-heavy engineering framework for factor discovery using a DSL and deterministic evaluation, rather than deriving novel mathematical proofs. While it touches on symbolic search, the core value proposition is the empirical workflow and detailed out-of-sample performance validation in cryptocurrency markets.

Research Flowchart

  flowchart TD
    A[Research Goal: Reproducible Factor Discovery] --> B[Input: Crypto Market Data & Experiment Trace]
    B --> C[Constraint: Point-in-time Factor DSL]
    C --> D[Process: Sequential Hypothesis Generation & Testing]
    D --> E[Engine: Deterministic Selection & Portfolio Simulation]
    E --> F[Outcome: Auditable Factors & Superior Out-of-Sample Returns]

Cited References (197)

  1. Closing the GARCH gap: Continuous time GARCH modeling (1996)
  2. Cross-Stock Predictability via LLM-Augmented Semantic Networks — → 2604.19476
  3. Cryptocurrency momentum has (not) its moments (2025)
  4. Technical trading rules in the cryptocurrency market (2020)
  5. Jumps and stochastic volatility: Exchange rate processes implicit in deutsche mark options (1996)
  6. Pricing and calibration in the 4-factor path-dependent volatility model — → 2406.02319
  7. The risk premia embedded in index options (2015)
  8. Model specification and risk premia: Evidence from futures options (2007)
and 189 more
  1. Forecasting with option-implied information (2013)
  2. Measuring tail risk (2024)
  3. Information in derivatives markets: Forecasting prices with prices (2025)
  4. Merton (1976) implied jump (2025)
  5. Empirical option pricing models (2022)
  6. Levy jump risk: Evidence from options and returns (2014)
  7. Post-‘87 crash fears in the S&P 500 futures option market (2000)
  8. Transform analysis and asset pricing for affine jump-diffusions (2000)
  9. Fast filtering with large option panels: Implications for asset pricing (2024)
  10. A tale of two option markets: Pricing kernels and volatility risk (2016)
  11. Specification analysis of option pricing models based on time-changed L 'e vy processes (2004)
  12. Smiles & smirks: Volatility and leverage by jumps (2022)
  13. Assessing models of individual equity option prices (2021)
  14. A closed-form solution for options with stochastic volatility with applications to bond and currency options (1993)
  15. Option valuation using the fast Fourier transform (1999)
  16. A novel pricing method for European options based on Fourier -cosine series expansions (2009)
  17. Efficient options pricing using the fast Fourier transform (2011)
  18. The dynamics of stochastic volatility: Evidence from underlying and options markets (2003)
  19. Alternative models for stock price dynamics (2003)
  20. Volatility dynamics for the S&P 500 : Further evidence from non-affine, multi-factor jump diffusions (2012)
  21. Empirical analysis of affine versus nonaffine variance specifications in jump-diffusion models for equity indices (2015)
  22. Jump and volatility dynamics for the S&P 500: Evidence for infinite-activity jumps with non-affine volatility dynamics from stock and option markets (2017)
  23. Model Complexity and Out-of-Sample Performance: Evidence from S&P 500 Index Returns (2018)
  24. Option profit and loss attribution and pricing: A new framework (2020)
  25. Interpretability in deep learning for finance: A case study for the Heston model (2026)
  26. Is technical analysis in the foreign exchange market profitable? A genetic programming approach (1997)
  27. Technical analysis as a sentiment barometer and the cross-section of stock returns (2023)
  28. AI Agent Misinformation when Assisting Financial Decision-Making: Early Evidence From Stock Recommendations (2025)
  29. Deep learning volatility: a deep neural network perspective on pricing and calibration in (rough) volatility models (2021)
  30. On Options-Driven Realized Volatility Forecasting: Information Gains via Rough Volatility Model (2025)
  31. Large language models for financial and investment management: Applications and benchmarks (2024)
  32. Large Language Models for Financial and Investment Management: Models, Opportunities, and Challenges. (2024)
  33. Expected returns and large language models (2022)
  34. A simple, positive semi-definite, heteroskedasticity and autocorrelationconsistent covariance matrix (1987)
  35. Technical indicators and the cross-section of corporate bond returns in a machine learning era (2025)
  36. ChatGPT and DeepSeek : Can they predict the stock market and macroeconomy? (2025) — link
  37. Generative AI for Finance: A New Framework (2025)
  38. Option valuation under no-arbitrage constraints with neural networks (2021)
  39. A nonparametric approach to pricing and hedging derivative securities via learning networks (1994)
  40. Deep Learning (2016)
  41. Forecasting value-at-risk using deep neural network quantile regression (2024)
  42. Inferring volatility dynamics and risk premia from the S&P 500 and VIX markets (2019)
  43. Explainable machine learning models of consumer credit risk (2023)
  44. Increasing the transparency of pricing dynamics in the US commercial real estate market with interpretable machine learning algorithms (2023)
  45. Explainable AI in credit risk management (2021) — link
  46. On deep calibration of (rough) stochastic volatility models (2025)
  47. Deep Calibration with Artificial Neural Network: A Performance Comparison on Option-Pricing Models (2023)
  48. A new closed-form discrete-time option pricing model with stochastic volatility (2023)
  49. Pricing and calibration in the 4-factor path-dependent volatility model — → 2406.02319
  50. Forecasting gold price with the XGBoost algorithm and SHAP interaction values (2021)
  51. Bankruptcy prediction using machine learning and Shapley additive explanations (2023)
  52. XAI –explainable artificial intelligence (2019)
  53. Approximation rates for deep calibration of (rough) stochastic volatility models (2024)
  54. Machine Learning and Data Sciences for Financial Markets: A Guide to Contemporary Practices (2023)
  55. Statistical machine learning for quantitative finance (2023)
  56. Explainable Artificial Intelligence (XAI) : Concepts, taxonomies, opportunities and challenges toward responsible AI (2020)
  57. Deep calibration of the quadratic rough Heston model (2022)
  58. Machine learning methods in finance: Recent applications and prospects (2023)
  59. ( R e-) Imag (in) ing price trends (2023)
  60. Narrative asset pricing: Interpretable systematic risk factors from news text (2023)
  61. Large Language Models and Futures Price Factors in China (2026)
  62. A taxonomy of anomalies and their trading costs (2016)
  63. Foundations of technical analysis: Computational algorithms, statistical inference, and empirical implementation (2000)
  64. When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks (2011)
  65. Can ChatGPT forecast stock price movements? return predictability and large language models — → 2304.07619
  66. Characteristics are covariances: A unified model of risk and return (2019)
  67. Returns to buying winners and selling losers: Implications for stock market efficiency (1993)
  68. On persistence in mutual fund performance (1997)
  69. Short-Term Moving Average Distance and the Cross-Section of Stock Returns (2025)
  70. False (and missed) discoveries in financial economics (2020)
  71. Lucky factors (2021)
  72. Advances in financial machine learning (2018)
  73. Implementing AI Foundation Models in Asset Management: A Practical Guide. (2025)
  74. Systematic investment strategies (2017)
  75. Chain-of-thought prompting elicits reasoning in large language models (2022)
  76. Choices matter when training machine learning models for return prediction (2024)
  77. Beyond Fama-French factors: Alpha from short-term signals (2023)
  78. Reflexion: Language agents with verbal reinforcement learning (2023)
  79. ReAct : Synergizing Reasoning and Acting in Language Models (2023)
  80. On Cross-Stock Predictability of Peer Return Gaps in China (2026)
  81. Industry return predictability: A machine learning approach (2019)
  82. Alpha go everywhere: Machine learning and international stock returns (2025)
  83. Empirical investigation of an equity pairs trading strategy (2019)
  84. Common risk factors in the returns on stocks and bonds (1993)
  85. Simple technical trading rules and the stochastic properties of stock returns (1992)
  86. Persuading investors: A video-based study (forthcoming)
  87. Autonomous Market Intelligence: Agentic AI Nowcasting Predicts Stock Returns (2026)
  88. Capital asset prices: A theory of market equilibrium under conditions of risk (1964)
  89. Neural network-based automatic factor construction (2020)
  90. Selecting Computational Models for Asset Management: Financial Econometrics versus Machine Learning-Is There a Conflict? (2020)
  91. … and the cross-section of expected returns (2016)
  92. A trend factor: Any economic gains from using information over investment horizons? (2016)
  93. A new anomaly: The cross-sectional profitability of technical analysis (2013)
  94. The virtue of complexity in return prediction (2024)
  95. Financial machine learning (2023)
  96. Machine learning in energy economics and finance: A review (2019)
  97. Machine learning in accounting and finance research: A literature review (2024)
  98. RESHAPE: Explaining accounting anomalies in financial statement audits by enhancing SHapley additive explanations (2022)
  99. Machine learning for quantitative finance: Fast derivative pricing, hedging and fitting (2018)
  100. Deep Learning for Exotic Option Valuation (2022)
  101. A neural network approach to understanding implied volatility movements (2020)
  102. Wavelet-based option pricing: An empirical study (2019)
  103. Deep learning of transition probability densities for stochastic asset models with applications in option pricing (2025)
  104. Stripping the Discount Curve–a Robust Machine Learning Approach (2026)
  105. Deep learning in asset pricing (2024)
  106. Deep learning statistical arbitrage (2026)
  107. Deep neural networks, gradient-boosted trees, random forests: Statistical arbitrage on the S&P 500 (2017)
  108. Industry return prediction via interpretable deep learning (2025)
  109. Taming the curse of dimensionality: quantitative economics with deep learning (2024)
  110. Short-term market risks implied by weekly options (2017)
  111. Parametric inference and dynamic state recovery from option panels (2015)
  112. How and when are high-frequency stock returns predictable? (2026)
  113. Intraday market return predictability culled from the factor zoo (2025)
  114. A Value for $n$-person Games (1953) — link
  115. A unified approach to interpreting model predictions (2017)
  116. Mathematical methods for financial markets (2009)
  117. On the nature of (jump) skewness risk premia (2024)
  118. The jump-risk premia implicit in options: Evidence from an integrated time-series study (2002)
  119. Volatility Models in Practice: Rough, Path-Dependent, or Markovian ? (2025)
  120. Joint SPX & VIX calibration with Gaussian polynomial volatility models: Deep pricing with quantization hints (2025)
  121. The price of the smile and variance risk premia (2021)
  122. Riding on the smiles (2011)
  123. Persistent and transient variance components in option pricing models with variance-dependent Kernel (2024)
  124. A First Option Calibration of the GARCH Diffusion Model by a PDE Method (2018) — link
  125. Switching to nonaffine stochastic volatility: A closed-form expansion for the inverse gamma model (2016)
  126. Time-series and cross-sectional stock return forecasting: New machine learning methods (2020)
  127. Artificial Intelligence–Powered (Finance) Scholarship (2026)
  128. Beyond prompting: An autonomous framework for systematic factor investing via agentic ai — → 2603.14288
  129. Social media-based attention and the cross-section of cryptocurrency returns (2025)
  130. Cryptocurrency trading: a comprehensive survey (2022)
  131. One News, Two Markets: LLM-Derived Sentiment and Trading Volume (2025)
  132. Behavioral anomalies in cryptocurrency markets (2019)
  133. Predicting cryptocurrency returns with machine learning: Evidence from high-dimensional factor modeling (2025)
  134. State-Dependent Relationship Between Cryptocurrency Returns and Credit Spreads (2026)
  135. Common risk factors in cryptocurrency (2022)
  136. Risks and returns of cryptocurrency (2021)
  137. A Trend Factor for the Cross Section of Cryptocurrency Returns (2025)
  138. Empirical Crypto Asset Pricing (2024) — link
  139. Cryptocurrency Volatility Forecasting with Applications in Trading (2025)
  140. Volatility forecasting with machine learning and intraday commonality (2024)
  141. Factor models, machine learning, and asset pricing (2022)
  142. Presidential address: Discount rates (2011)
  143. Automated volatility forecasting (2025)
  144. Empirical Dynamic Asset Pricing: Model Specification and Econometric Assessment (2006)
  145. Empirical performance of alternative option pricing models (1997)
  146. A New Model for the Joint Valuation of S&P 500 and VIX Options: Specification Analysis (2025)
  147. VIX derivatives, hedging and vol-of-vol risk (2020)
  148. The quadratic rough Heston model and the joint S&P 500/VIX smile calibration problem (2020)
  149. 4/2 rough and smooth (2025)
  150. The volatility surface: A Practitioner’s Guide (2011)
  151. Model complexity and out-of-sample performance: Evidence from S&P 500 index returns (2018)
  152. Cross-stock momentum and factor momentum (2023)
  153. How Much of Cross-Stock Momentum Reflects Underreaction? (2025)
  154. Can ChatGPT forecast stock price movements? Return predictability and large language models (2023)
  155. Expected returns and large language models (2023)
  156. Bloomberg GPT : A large language model for finance (2023) — link
  157. From Text to Alpha: Can LLM s Track Evolving Signals in Corporate Disclosures? (2025) — link
  158. Finben: A holistic financial benchmark for large language models (2024)
  159. Structured Beliefs and Fund Performance: An LLM -Based Approach (2024)
  160. Pixiu: A large language model, instruction data and evaluation benchmark for finance (2023) — link
  161. A Financial Brain Scan of the LLM — → 2508.21285
  162. Dual Peer Effects and Cross-Stock Predictability (2026)
  163. On cross-stock predictability of peer return gaps in China (2026)
  164. Economic links and predictable returns (2008)
  165. Do industries explain momentum? (1999)
  166. Geographic lead-lag effects (2020)
  167. Shared analyst coverage: Unifying momentum spillover effects (2020)
  168. Principal portfolios (2023)
  169. Text-based industry momentum (2018)
  170. Market segmentation and cross-predictability of returns (2010)
  171. Modeling the Implied Volatility Smirk in China : Do Non-Affine Two-Factor Stochastic Volatility Models Work? (2025)
  172. Chronologically consistent large language models — → 2502.21206
  173. Machine learning and the stock market (2023)
  174. Deep surrogate for non-affine stochastic volatility option valuation models (2026)
  175. Trading Volume Alpha (2024)
  176. Presidential address: The scientific outlook in financial economics (2017)
  177. Giving content to investor sentiment: The role of media in the stock market (2007)
  178. Empirical asset pricing via machine learning (2020)
  179. A five-factor asset pricing model (2015)
  180. GPT ’s idea of stock factors (2024)
  181. Agentic AI and the Future of Institutional Asset Management. (2025)
  182. Machine learning vs. economic restrictions: Evidence from stock return predictability (2023)
  183. Pairs trading: Performance of a relative-value arbitrage rule (2006)
  184. Network momentum across asset classes — → 2308.11294
  185. Production complementarity and information transmission across industries (2024)
  186. Technological links and predictable returns (2019)
  187. The effect of innovation similarity on asset prices: Evidence from patents’ big data (2023)
  188. Competition links and stock returns (2022)
  189. The Memorization Problem: Can We Trust LLM s’ Economic Forecasts? — → 2504.14765

9 of 197 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.