Paper: arXiv 2508.02366

Authors: Adam Darmanin, Vince Vella

Abstract

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 reasoning and multi-modal signal interpretation when guided by well-structured prompts. This paper proposes a hybrid framework in which LLMs generate high-level trading strategies to guide RL agents. We evaluate (i) the economic rationale of LLM-generated strategies through expert review, and (ii) the performance of LLM-guided agents against unguided RL baselines using Sharpe Ratio (SR) and Maximum Drawdown (MDD). Empirical results indicate that LLM guidance improves both return and risk metrics relative to standard RL.

Complexity vs Empirical Score

  • Math Complexity: 6.5/10
  • Empirical Rigor: 6.0/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper employs advanced reinforcement learning algorithms, hierarchical prompting frameworks, and regret minimization heuristics for prompt tuning, indicating significant mathematical complexity; it also provides empirical backtesting using standardized financial metrics (Sharpe Ratio, Max Drawdown) on real data, with a structured benchmark environment and reproducible methodology.

Research Flowchart

  flowchart TD
  A["Research Goal: Address myopic, opaque RL policies in quantitative trading using LLM guidance"] --> B["Methodology: Hybrid Framework LLM generates high-level strategy guiding RL agent"]
  B --> C["Data/Inputs: Multi-modal market signals, financial objectives, expert review criteria"]
  C --> D["Computational Process 1: LLM prompt analysis & strategy generation"]
  D --> E["Computational Process 2: RL agent learning guided by LLM strategy"]
  E --> F["Computational Process 3: Performance evaluation vs. baseline RL"]
  F --> G["Key Findings/Outcomes: Expert validation of economic rationale; Improved SR & MDD vs. unguided RL"]

Cited References (95)

  1. IEEE tran Homepage (2007) — link
  2. IEEE tran Webpage on CTAN (2007) — link
  3. The IEEE Website (2007) — link
  4. \BibTeX ing (1988) — link
  5. Designing \BibTeX Styles (1988) — link
  6. Tame the BeaST — The B to X of \BibTeX (2005) — link
  7. \BibTeX Tips and FAQ (2007) — link
  8. The \TeX FAQ (2007) — link
and 87 more
  1. A \BibTeX Guide via Examples (2004) — link
  2. \TeX User Group Bibliography Archive (2006) — link
  3. The natbib.sty package (2006) — link
  4. The url.sty Package (2005) — link
  5. The hyperref.sty Package (2006) — link
  6. The breakurl.sty Package (2006) — link
  7. The Babel Package (2005) — link
  8. The multibib.sty package (2004) — link
  9. The biblatex package (2007) — link
  10. A Novel Ultrathin Elevated Channel Low-temperature Poly- Si TFT (1999)
  11. Butt-jointed DBR Laser With 15 nm Tunability Grown in Three MOVPE Steps (1995)
  12. Pull-in Time Dynamics as a Measure of Absolute Pressure (1997)
  13. Introduction to Magnetic Materials (1972)
  14. Midgap Traps Related to Compensation Processes in CdTe Alloys (1997)
  15. Analysis and Comparison of PM Synchronous Motor and Induction Motor Type Magnetic Bearings (1995)
  16. Internet Tomography (2002)
  17. On the Minimum Distance of Parallel and Serially Concatenated Codes (submitted for publication)
  18. Oversampling Delta-Sigma Data Converters Theory, Design and Simulation (1992)
  19. Laser Assisted Microtechnology (1998)
  20. The Analysis of Directional Time Series: Applications to Wind Speed and Direction (1989)
  21. A Course in Number Theory (1988)
  22. Theory Principles and Design of Magnetic Circuits (1964)
  23. Optical Reflectometry for Component Characterization (1998)
  24. Properties of the Tailbiting BCJR Decoder (2000)
  25. Theory for Transmission of Vector Quantization Data (1995)
  26. Design of an Improved Pixel for a Polysilicon Active-Matrix Organic LED Display (1998)
  27. FLEXChip Signal Processor ( MC68175/D ) (1996)
  28. FLEXChip Signal Processor (1996)
  29. Radial Basis Function Decision Feedback Equaliser Assisted Burst-by-burst Adaptive Modulation (1999)
  30. Measurement and Modeling of the Temporal Dependence in Packet Loss (1999)
  31. High Resolution Fiber Distributed Measurements With Coherent OFDR (2000)
  32. 160 Gbit/s Single-channel Transmission Over 300 km Nonzero-dispersion Fiber With Semiconductor Based Transmitter and Demultiplexer (1999)
  33. A Novel Approach to Automatic Protection Switching Using Trees (1997)
  34. High-Resolution Micromachined Interferometric Accelerometer (1992)
  35. Performance of TCP Congestion Control with Rate Feedback: TCP/ABR and Rate Adaptive TCP/IP (1999)
  36. Delay Characterization and Performance Control of Wide-area Networks (2000) — link
  37. Congestion Avoidance in Computer Networks with a Connectionless Network Layer (1987)
  38. A Stochastic Model of TCP R eno Congestion Avoidance and Control (1999)
  39. A Tutorial Review of Elements of Weak Signal Detection in Non- G aussian EMI Environments (1986)
  40. TCP over ATM : ABR or UBR
  41. Modified TCP Congestion Avoidance Algorithm (1990) — link
  42. Cache Array Routing Protocol v1.1 (1998) — link
  43. Optimal Partition of QoS Requirements on Unicast Paths and Multicast Trees (1998) — link
  44. High-Speed Digital-to- RF Converter (1997)
  45. Quadrature Modulation Circuit (1992)
  46. Estimateur d’un D 'e faut de Fonctionnement d’un Modulateur en Quadrature et 'E tage de Modulation l’Utilisant
  47. IEEE_M_PCOM # “, Special Issue on Wireless ATM " (1996)
  48. Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specification (1997)
  49. Fiber Channel Physical Interface ( FC-PI ) (1999)
  50. MATE : MPLS Adaptive Traffic Engineering (1999)
  51. Enhanced Proportional Rate Control Algorithm PRCA (1994)
  52. Advanced QoS Services for the Intelligent Internet (1997)
  53. PDCA12-70 Data Sheet
  54. A Proposal to Add Explicit Congestion Notification ( ECN ) to IP (1999)
  55. SaberDesigner Reference Manual (1998)
  56. Ucb/lbnl/vint Network Simulator—ns (Version 2) — link
  57. M essung von S t "o rfeldern an A nlagen und L eitungen der T elekommunikation im F requenzbereich 9 kHz bis 3 GHz
  58. Telemetry Channel Coding (1999) — link
  59. Telemetry Channel Coding (1999) — link
  60. Telemetry Channel Coding (1999) — link
  61. Telemetry Channel Coding (1999) — link
  62. Deep Reinforcement Learning: A Brief Survey (2017) — link
  63. Deep Reinforcement Learning with Double Q-Learning (2016) — link
  64. Reinforcement Learning: An Introduction (2018) — link
  65. An Application of Deep Reinforcement Learning to Algorithmic Trading (2021) — link
  66. Deep Reinforcement Learning Applied to a Sparse-Reward Trading Environment with Intraday Data (2024) — link
  67. FinRL: Deep Reinforcement Learning Framework to Automate Trading in Quantitative Finance (2021) — link
  68. Reinforcement Learning for Quantitative Trading (2023) — link
  69. Beyond Econometrics: A Roadmap Towards Financial Machine Learning (2019) — link
  70. The 10 Reasons Most Machine Learning Funds Fail (2018) — link
  71. FinMem: A Performance-Enhanced LLM Trading Agent with Layered Memory and Character Design (2024) — link
  72. A Multimodal Foundation Agent for Financial Trading: Tool-Augmented, Diversified, and Generalist (2024) — link
  73. Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models (2023)
  74. The Memorization Problem: Can We Trust LLMs’ Economic Forecasts? (2025) — link
  75. Leveraging LLMs for Financial News Analysis and Macroeconomic Indicator Nowcasting (2024) — link
  76. LLMFactor : Extracting Profitable Factors through Prompts for Explainable Stock Movement Prediction (2024) — link
  77. QuantAgent: Seeking Holy Grail in Trading by Self-Improving Large Language Model (2024) — link
  78. Demystifying Prompts in Language Models via Perplexity Estimation (2023) — link
  79. The Prompt Report: A Systematic Survey of Prompting Techniques (2024) — link
  80. Towards Reasoning in Large Language Models: A Survey (2022) — link
  81. Can Large Language Models Faithfully Express Their Intrinsic Uncertainty in Words? (2024) — link
  82. Entropy Heat-Mapping: Localizing GPT -Based OCR Errors with Sliding-Window Shannon Analysis (2025) — link
  83. Artificial Intelligence in Asset Management (2020)
  84. A Five-Factor Asset Pricing Model (2015) — link
  85. Examining the Predictive Power of Moving Averages in the Stock Market (2022) — link
  86. High-frequency trading: Better than its reputation? (2011) — link
  87. Explainable AI for Interpretable Credit Scoring (2020) — link

0 of 95 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.