arXiv cs.AI· Jie Huang, Yuecheng Hong, Yao Chen, Hao Zhong, Linfeng Xiao, Xudong Liu, Qiang Ye, Hao Xia, Jianxuan Xin·· 3 天前AI 评分24
AlphaRLM：让语言模型从文献出发发现符号化因子
AlphaRLM: Aligning Language Models for Literature-Grounded Symbolic Factor Discovery
AI 导读
AlphaRLM 是一个强化学习框架，训练 LLM 直接从金融文献构建可执行的符号化因子，无需输入因子表达式。它通过文献引导的算子组装（LEGO）提供语义先验，并用可执行有效性与分组收益分离奖励对齐市场反馈。在 CSI 300、CSI 1000 和 S&P 500 上，其年化超额收益优于强因子挖掘与 LLM 基线，8B 模型展现出较强的跨市场表现。
正文 · 原文
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.
| Comments: | 25 pages, 5 figures |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG); Portfolio Management (q-fin.PM) |
| Cite as: | arXiv:2610.14959 [cs.AI] |
| (or arXiv:2610.14959v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.14959 arXiv-issued DOI via DataCite |
Submission history
From: Jie Huang [view email]
[v1]
Sat, 10 Oct 2026 14:32:10 UTC (2,914 KB)
[v2]
Wed, 14 Oct 2026 12:33:55 UTC (2,914 KB)
来源：arXiv cs.AI · arxiv.org