AlphaRLM: Aligning Language Models for Literature-Grounded Symbolic Factor Discovery

By Jie Huang, Yuecheng Hong, Yao Chen, Hao Zhong, Linfeng Xiao, Xudong Liu, Hao Xia, Jianxuan Xin

Rating

1828
Battle Count: 50

Relevance

10/10
The paper directly addresses alpha factor discovery, a core component of quantitative trading. It proposes a novel method for generating executable trading signals from literature and validates it through rigorous backtesting on major equity indices.

Implementation Complexity

8/10
High complexity due to the integration of LLM fine-tuning (DAPO), symbolic expression parsing/validation, and market-based reward calculation. Requires significant computational resources (8x A100 GPUs) and expertise in both NLP and quantitative finance.

Reproducibility

3/5
The paper provides detailed hyperparameters, prompt templates, and data split descriptions. However, no public code repository URL is explicitly provided in the text. The literature corpus construction involves proprietary filtering steps (GPT-4 relevance) and specific journal sources, though the methodology is described in detail.

Abstract

Formulaic alpha mining seeks to discover executable expressions that capture predictive market signals in a vast, noisy search space. Recent LLM-based approaches use prompting, iterative search, or seed-factor evolution, leaving direct factor construction through policy learning less explored. To address this, we introduce AlphaRLM, a reinforcement-learning framework that trains LLMs to construct executable symbolic factors from financial literature without input factor expressions. AlphaRLM incorporates Literature-Guided Operator Assembly (LEGO), which conditions factor construction on literature-derived ideas and executable operators, providing semantic priors for exploration. It further aligns the generation policy with market feedback through executable-validity and grouped-return-separation rewards that capture non-monotonic structures across factor quantiles. Extensive experiments on CSI 300, CSI 1000, and S&P 500 show that AlphaRLM outperforms strong factor-mining and LLM-based baselines in annualized excess return. Notably, the 8B model achieves strong cross-market performance, highlighting the promise of literature-conditioned, market-aligned factor generation.

Links

arXiv:2610.14959 ยท Paper rankings

The interactive Everscope explorer (charts, battles, favorites) loads below.