Paper: arXiv 2512.23515

Authors: Zuoyou Jiang, Li Zhao, Rui Sun, Ruohan Sun, Zhongjian Li, Jing Li, Daxin Jiang, Zuo Bai, Cheng Hua

Abstract

Signal decay and regime shifts pose recurring challenges for data-driven investment strategies in non-stationary markets. Conventional time-series and machine learning approaches, which rely primarily on historical correlations, often struggle to generalize when the economic environment changes. While large language models (LLMs) offer strong capabilities for processing unstructured information, their potential to support quantitative factor screening through explicit economic reasoning remains underexplored. Existing factor-based methods typically reduce alphas to numerical time series, overlooking the semantic rationale that determines when a factor is economically relevant. We propose Alpha-R1, an 8B-parameter reasoning model trained via reinforcement learning for context-aware alpha screening. Alpha-R1 reasons over factor logic and real-time news to evaluate alpha relevance under changing market conditions, selectively activating or deactivating factors based on contextual consistency. Empirical results across multiple asset pools show that Alpha-R1 consistently outperforms benchmark strategies and exhibits improved robustness to alpha decay. The full implementation and resources are available at https://github.com/FinStep-AI/Alpha-R1.

Complexity vs Empirical Score

  • Math Complexity: 6.0/10
  • Empirical Rigor: 8.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper integrates reinforcement learning and advanced LLM reasoning with a substantial mathematical foundation, including statistical concepts for non-stationary markets and RL optimization. Empirically, it presents extensive backtesting across multiple asset pools, provides a public GitHub repository with implementation, and reports robustness to alpha decay, indicating high practical applicability.

Research Flowchart

  flowchart TD
  A["Research Goal<br>Address Signal Decay & Regime Shifts"] --> B["Data Inputs<br>Market Data, Unstructured News, Factor Series"]
  B --> C["Methodology<br>Alpha-R1 8B Model with RL Training"]
  C --> D["Process<br>Reasoning: Factor Logic + News Context"]
  D --> E["Outcome<br>Context-Aware Alpha Screening<br>(Activate/Deactivate Factors)"]
  E --> F["Key Findings<br>Outperforms Benchmarks, Robust to Decay"]

Cited References (90)

  1. Reinforcement-Learning Portfolio Allocation with Dynamic Embedding of Market Information — → 2501.17992
  2. 101 formulaic alphas (2016) — link
  3. DeepSeekMath : Pushing the Limits of Mathematical Reasoning in Open Language Models (2024) — link
  4. Common risk factors in the returns on stocks and bonds (1993) — link
  5. Capital Asset Prices: A Theory of Market Equilibrium under Conditions of Risk (1964) — link
  6. On Persistence in Mutual Fund Performance (1997) — link
  7. A Five-Factor Asset Pricing Model (2015) — link
  8. Taming the Factor Zoo: A Test of New Factors (2020) — link
and 82 more
  1. Can ChatGPT Forecast Stock Price Movements? Return Predictability and Large Language Models (2026) — link
  2. Dissecting Characteristics Nonparametrically (2020) — link
  3. Quantum Finance and Fuzzy Reinforcement Learning-Based Multi-Agent Trading System (2024) — link
  4. Alpha-GPT 2.0: Human-in-the-Loop AI for Quantitative Investment (2024) — link
  5. R&D-Agent-Quant : A Multi-Agent Framework for Data-Centric Factors and Model Joint Optimization (2025) — link
  6. Automate Strategy Finding with LLM in Quant Investment — → 2409.06289
  7. AlphaAgent : LLM -Driven Alpha Mining with Regularized Exploration to Counteract Alpha Decay (2025) — link
  8. AlphaForge : a framework to mine and dynamically combine formulaic alpha factors (2025) — link
  9. AlphaEval : A Comprehensive and Efficient Evaluation Framework for Formula Alpha Mining (2026) — link
  10. Fine-Tuning Large Language Models for Stock Return Prediction Using Newsflow — → 2407.18103
  11. From Deep Learning to LLMs : A Survey of AI in Quantitative Investment (2025) — link
  12. Alpha- GPT : Human- AI Interactive Alpha Mining for Quantitative Investment (2025) — link
  13. Regression Shrinkage and Selection via the Lasso (1996) — link
  14. Factor Investment or Feature Selection Analysis? (2025) — link
  15. Empirical Asset Pricing via Machine Learning (2020) — link
  16. Factor Models, Machine Learning, and Asset Pricing (2022) — link
  17. A Survey of Reinforcement Learning from Human Feedback (2025) — link
  18. Training language models to follow instructions with human feedback (2022) — link
  19. Proximal Policy Optimization Algorithms (2017) — link
  20. DeepSeek - R1 incentivizes reasoning in LLMs through reinforcement learning (2025) — link
  21. BloombergGPT : A Large Language Model for Finance (2023) — link
  22. Fin GPT : Democratizing Internet-scale Data for Financial Large Language Models (2023) — link
  23. Instruct-FinGPT : Financial Sentiment Analysis by Instruction Tuning of General-Purpose Large Language Models (2023) — link
  24. Lo RA : Low-Rank Adaptation of Large Language Models (2022) — link
  25. PIXIU : A Comprehensive Benchmark, Instruction Dataset and Large Language Model for Finance (2023) — link
  26. BBT-Fin : Comprehensive Construction of Chinese Financial Domain Pre-trained Language Model, Corpus and Benchmark (2023) — link
  27. XuanYuan 2.0 : A Large Chinese Financial Chat Model with Hundreds of Billions Parameters (2023) — link
  28. Trading-R1 : Financial Trading with LLM Reasoning via Reinforcement Learning (2025) — link
  29. Fin-R1 : A Large Language Model for Financial Reasoning through Reinforcement Learning (2025) — link
  30. FinO1 : On the Transferability of Reasoning-Enhanced LLMs and Reinforcement Learning to Finance (2025) — link
  31. TradingGPT : Multi-Agent System with Layered Memory and Distinct Characters for Enhanced Financial Trading Performance (2023) — link
  32. FinMem : A Performance-Enhanced LLM Trading Agent with Layered Memory and Character Design (2025) — link
  33. TradingAgents : Multi-Agents LLM Financial Trading Framework (2025) — link
  34. MetaGPT : Meta Programming for A Multi-Agent Collaborative Framework (2024) — link
  35. A Multimodal Foundation Agent for Financial Trading: Tool-Augmented, Diversified, and Generalist (2024) — link
  36. Learning to Generate Explainable Stock Predictions using Self-Reflective Large Language Models — → 2402.03659
  37. LLMF actor: Extracting Profitable Factors through Prompts for Explainable Stock Movement Prediction (2024) — link
  38. Direct Preference Optimization: Your Language Model is Secretly a Reward Model (2023) — link
  39. SimPO : Simple Preference Optimization with a Reference-Free Reward (2024) — link
  40. Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback (2023) — link
  41. R eward B ench: Evaluating Reward Models for Language Modeling (2025) — link
  42. Measuring Massive Multitask Language Understanding (2021) — link
  43. Towards Mitigating LLM Hallucination via Self Reflection (2023) — link
  44. Beyond the Black Box: Interpretability of LLMs in Finance (2025) — link
  45. FinRL : A Deep Reinforcement Learning Library for Automated Stock Trading in Quantitative Finance (2020) — link
  46. Continuous control with deep reinforcement learning (2016) — link
  47. Addressing Function Approximation Error in Actor-Critic Methods (2018)
  48. Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor (2018)
  49. Stable-Baselines3 : Reliable Reinforcement Learning Implementations (2021)
  50. MAPS : Multi-agent Reinforcement Learning-based Portfolio Management System (2020) — link
  51. Reinforcement-Learning Based Portfolio Management with Augmented Asset Movement Prediction States (2020) — link
  52. Chain-of-Alpha : Unleashing the Power of Large Language Models for Alpha Mining in Quantitative Trading (2025) — link
  53. Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency (1993)
  54. Evidence of Predictable Behavior of Security Returns (1990)
  55. The Deflated Sharpe Ratio: Correcting for Selection Bias, Backtest Overfitting and Non-Normality (2014)
  56. The Probability of Backtest Overfitting (2017)
  57. The Sharpe Ratio Efficient Frontier (2012)
  58. The Stationary Bootstrap (1994)
  59. A Simple, Positive Semi-Definite, Heteroskedasticity and Autocorrelation Consistent Covariance Matrix (1987)
  60. The Probabilistic Relevance Framework: BM25 and Beyond (2009)
  61. AlphaAgentEvo : Evolution-Oriented Alpha Mining via Self-Evolving Agentic Reinforcement Learning (2026) — link
  62. Navigating the Alpha Jungle: An LLM -Powered MCTS Framework for Formulaic Alpha Factor Mining (2026) — link
  63. QuantaAlpha : An Evolutionary Framework for LLM -Driven Alpha Mining (2026) — link
  64. FactorMiner : A Self-Evolving Agent with Skills and Experience Memory for Financial Alpha Discovery (2026) — link
  65. AlphaCrafter : Harnessing Multi-Agent Workflows for Cross-Sectional Quantitative Trading (2026) — link
  66. From Feedback Loops to Policy Updates: Reinforcement Fine-Tuning for LLM -Based Alpha Factor Discovery (2026) — link
  67. Can LLM -based Financial Investing Strategies Outperform the Market in Long Run? (2026) — link
  68. Time Travel is Cheating: Going Live with DeepFund for Real-Time Fund Investment Benchmarking (2025) — link
  69. QuantEvolve : Automating Quantitative Strategy Discovery through Multi-Agent Evolutionary Framework (2025) — link
  70. AlphaAgents : Large Language Model based Multi-Agents for Equity Portfolio Constructions (2025) — link
  71. AlphaQuanter : An End-to-End Tool-Augmented Agentic Reinforcement Learning Framework for Stock Trading (2026) — link
  72. Trade- R1 : Bridging Verifiable Rewards to Stochastic Environments via Process-Level Reasoning Verification (2026) — link
  73. Reasoning through Verifiable Forecast Actions: Consistency-Grounded RL for Financial LLMs (2026) — link
  74. FinWorld : An All-in-One Open-Source Platform for End-to-End Financial AI Research and Deployment (2026) — link
  75. Understanding R1 -Zero-Like Training: A Critical Perspective (2025) — link
  76. DAPO : An Open-Source LLM Reinforcement Learning System at Scale (2025) — link
  77. Group Sequence Policy Optimization (2025) — link
  78. The Entropy Mechanism of Reinforcement Learning for Reasoning Language Models (2025) — link
  79. Forest through the Trees: Building Cross-Sections of Stock Returns (2025) — link
  80. Growing the Efficient Frontier on Panel Trees (2025) — link
  81. APT or `` AIPT ‘’? The Surprising Dominance of Large Factor Models (2024) — link
  82. Structural Deep Learning in Conditional Asset Pricing (2022) — link

4 of 90 cited works are in the corpus; resolved via arXiv source bundles, Semantic Scholar and OpenAlex. Corpus-wide: 2518 linked papers.