Trading-R1: Financial Trading with LLM Reasoning via Reinforcement Learning

Yijia Xiao, Edward Sun, Tong Chen, Fang Wu, Di Luo +1 more
9/14/2025
q-fin.TRcs.AIcs.CE

No official code found. Best implementation: co1onnese/NeMo-Charlie (2 stars, last commit 2025-10-28). 4 more below.

Abstract

Developing professional, structured reasoning on par with human financial analysts and traders remains a central challenge in AI for finance, where markets demand interpretability and trust. Traditional time-series models lack explainability, while LLMs face challenges in turning natural-language analysis into disciplined, executable trades. Although reasoning LLMs have advanced in step-by-step planning and verification, their application to risk-sensitive financial decisions is underexplored. We present Trading-R1, a financially-aware model that incorporates strategic thinking and planning for comprehensive thesis composition, facts-grounded analysis, and volatility-adjusted decision making. Trading-R1 aligns reasoning with trading principles through supervised fine-tuning and reinforcement learning with a three-stage easy-to-hard curriculum. Training uses Tauric-TR1-DB, a 100k-sample corpus spanning 18 months, 14 equities, and five heterogeneous financial data sources. Evaluated on six major equities and ETFs, Trading-R1 demonstrates improved risk-adjusted returns and lower drawdowns compared to both open-source and proprietary instruction-following models as well as reasoning models. The system generates structured, evidence-based investment theses that support disciplined and interpretable trading decisions. Trading-R1 Terminal will be released at https://github.com/TauricResearch/Trading-R1.

View on arXivView PDF

Code Implementations(5)

co1onnese/NeMo-CharlieCommunity56%

SFT & RFL experiments on DeepSeek-V3 inspired by the Tauric-TR1-DB presented in the research paper: Trading-R1: Financial Trading with LLM Reasoning via Reinforcement Learning -> https://arxiv.org/abs/2509.11420

21Shell, PythonOct 26, 202511 months ago
co1onnese/Charlie-T1-DBCommunity56%

a database for stock market analysis inspired by the Tauric-TR1-DB presented in the research paper: Trading-R1: Financial Trading with LLM Reasoning via Reinforcement Learning -> https://arxiv.org/abs/2509.11420

00PythonOct 22, 202512 months ago
co1onnese/Trainer-CharlieCommunity55%

LLM Tuning data preparation inspired by the Tauric-TR1-DB presented in the research paper: Trading-R1: Financial Trading with LLM Reasoning via Reinforcement Learning -> https://arxiv.org/abs/2509.11420

10Shell, PythonOct 25, 202512 months ago
co1onnese/SFT-CharlieCommunity54%

SFT experiment on DeepSeek-V2.5 inspired by the Tauric-TR1-DB presented in the research paper: Trading-R1: Financial Trading with LLM Reasoning via Reinforcement Learning -> https://arxiv.org/abs/2509.11420

00Shell, PythonOct 26, 202512 months ago
co1onnese/TradingCharlieCommunity52%

a data pipeline for stock market analysis inspired by the research paper: Trading-R1: Financial Trading with LLM Reasoning via Reinforcement Learning -> https://arxiv.org/abs/2509.11420

00Shell, PythonOct 19, 202512 months ago

Similar Papers

Fino1: On the Transferability of Reasoning-Enhanced LLMs and Reinforcement Learning to Finance

87% match1 repo802025

Fin-R1: A Large Language Model for Financial Reasoning through Reinforcement Learning

86% match1 repo8182025

Language Model Guided Reinforcement Learning in Quantitative Trading

85% match2025

PRUDEX-Compass: Towards Systematic Evaluation of Reinforcement Learning in Financial Markets

84% match2023

Trade-R1: Bridging Verifiable Rewards to Stochastic Environments via Process-Level Reasoning Verification

84% match2026

A Multimodal Foundation Agent for Financial Trading: Tool-Augmented, Diversified, and Generalist

84% match2024

Cite this paper

@article{xiao2025tradingr,
  title  = {Trading-R1: Financial Trading with LLM Reasoning via Reinforcement Learning},
  author = {Yijia Xiao and Edward Sun and Tong Chen and Fang Wu and Di Luo and Wei Wang},
  year   = {2025},
  eprint = {2509.11420},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2509.11420v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2509.11420.svg)](https://codeofpaper.com/paper/2509.11420)

Discussion