Lineage-Aware Search Control for LLM-Based Evolutionary Factor Mining

Xintong Liu; Chenkai Zhang; Yanzhao Ma; Peng Yang

q-fin.CPOct 21, 2026

Lineage-Aware Search Control for LLM-Based Evolutionary Factor Mining

Authors: Xintong Liu, Chenkai Zhang, Yanzhao Ma, Peng Yang

Organizations: Southern University of Science and Technology (SUSTech) Shenzhen, China

Abstract

Formulaic alpha factor mining searches for symbolic expressions that transform market data into predictive signals. Recent LLM-based evolutionary factor mining methods use large language models as semantic variation operators. Under the conventional LLM-EA paradigm, stochastic variation and fitness-based selection can leave the search weakly directed and concentrated around a narrow set of structurally similar factors. We argue that these limitations arise because current methods mainly organize search at the individual and population levels, leaving lineage-level control underused. A lineage connects successive factor variants through evolutionary descent. We propose LineageEvo, which uses lineage in two coupled forms of search control: (1) lineage-conditioned LLM variation, which distills Lineage-Scoped Search Experience into operator-specific priors and fuses them with global operator priors, providing persistent, lineage-relevant guidance for LLM generation; and (2) lineage-stratified search allocation, which balances factor quality against lineage concentration by distributing the search budget across lineages during evolutionary search. Together, these mechanisms make semantic variation more direction aware and prevent branches from dominating the evolution, thereby promoting the discovery of predictive and diverse factors. We evaluate LineageEvo on CSI 500 and S&P 500 in terms of factor predictiveness, portfolio performance, search efficiency, and lineage- and expression-level diversity. LineageEvo achieves the strongest predictive performance on both markets, while LSPS produces a more balanced lineage distribution and a structurally broader factor pool. The results support lineage as a useful intermediate unit for controlling LLM-based evolutionary factor search.

Figures & tables

Figure 1: Overview of LineageEvo . LSPS first selects a lineage and then selects parent factor(s); lineage-specific and global operator priors are subsequently retrieved and fused to condition LLM-based variation. Valid offspring are evaluated, and the feedback path updates the lineage DAG, prior states, active population, and elite archive.

	CSI 500					S&P 500
Method	IC $\uparrow$	ICIR $\uparrow$	AR $\uparrow$	IR $\uparrow$	MDD $\downarrow$	IC $\uparrow$	ICIR $\uparrow$	AR $\uparrow$	IR $\uparrow$	MDD $\downarrow$
Alpha158	0.0076	0.0787	15.50%	0.6866	31.62%	0.0081	0.0426	17.58%	0.7715	21.03%
GP	0.0074	0.0774	15.28%	0.6861	33.08%	0.0055	0.0575	21.38%	1.0958	20.55%
AlphaGen	0.0345	0.2215	15.04%	0.7176	36.31%	0.0099	0.0587	20.38%	0.9709	23.87%
AlphaSAGE	0.0230	0.1855	13.71%	0.7654	21.31%	0.0118	0.0668	18.32%	0.8539	21.62%
RD-Agent	0.0395	0.2253	14.14%	0.7199	29.48%	0.0094	0.0464	11.05%	0.7642	20.61%

Table 1: Main results on CSI 500 and S&P 500. Bold and underlined values denote the best and second-best results.

Figure 2: Cumulative return curves of different factor mining methods on CSI 500 during the test period.

Variant	IC $\uparrow$	ICIR $\uparrow$
w/o all priors	0.0384	0.1994
w/o lineage priors	0.0390	0.2041
w/o global priors	0.0495	0.2579
w/o mutation priors	0.0438	0.2673
w/o crossover priors	0.0445	0.2362
Full LineageEvo	0.0512	0.2740

Table 2: Ablation study of the search-prior components on CSI 500.

Method	IC $\uparrow$	ICIR $\uparrow$	Lineage Entropy $\uparrow$	AST Dist. $\uparrow$	Eff. Struct. $\uparrow$
w/o LSPS	0.0381	0.2415	0.405	0.509	37.89
LineageEvo	0.0512	0.2740	2.154	0.747	41.90

Table 3: Effect of LSPS on predictive performance and diversity on CSI 500. Higher values are better.

Figure 3: Search efficiency comparison on CSI 500. The x-axis denotes the number of valid evaluated factors, and the y-axis reports the top-5 mean-so-far absolute IC.

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 4: A same-lineage prior-reuse case. Event 1 replaces TsSkew with a close/high-return TsCov , triggering a structured prior rewrite. Event 2 receives this prior on a sibling branch, applies the same replacement while preserving its denominator, and strengthens the pattern from medium to high confidence. The dashed connector denotes prior transfer.

Component	Parameter	Setting
LLM	Backend	deepseek-v4-flash through an OpenAI-compatible client
LLM	Temperature / maximum tokens	$0.7$ / $8000$
LLM	Timeout / retries / retry wait	$120$ s / $5$ / $5$ s
Search	Valid-evaluation budget	$B=200$
Search	Initial lineage roots	$10$ LLM-generated seed factors
Search	Valid mutation / crossover slots per generation	$3$ / $2$

Appendix

Table 4: Main implementation and evaluation parameters.

Prompt	Trigger	Main runtime inputs	Required output
Seed generation	Population initialization	Market, stock universe, factor DSL, expression-length target, and existing or rejected seed expressions	One factor with hypothesis and metadata in JSON
Mutation	After selecting one parent	Parent expression and metrics, lineage and global mutation priors, search-control state, fusion decision, and retry feedback	{"factor": "..."}
Crossover	After selecting two parents	Primary and secondary expressions and metrics, their lineage priors, global crossover prior, fusion decision, and duplicate feedback	{"factor": "..."}
Prior rewriting	After an informative valid evolutionary event	Previous prior, parent and child evidence, expression difference, update trigger, and target JSON schema	A complete updated prior in JSON

Appendix

Table 5: Summary of the principal prompts used by LineageEvo .

Alpha Factor Discovery

See all 21 papers

From Feedback Loops to Policy Updates: Reinforcement Fine-Tuning for LLM-Based Alpha Factor Discovery
May 14, 2026Lingzhe Zhang, Tong Jia, Yunpeng Zhai +5RL for Language Models Alpha Factor Discovery
Peking University, Beijing, China · Alibaba Group, China · Nanjing University, Nanjing, China +1
AlphaSchema: Exploring the Space of Trading Semantics for LLM-Based Alpha Mining
Jul 29, 2026Jingyang Yi, Jian Yang, Yifei Jin +2Automated Algorithm Discovery Alpha Factor Discovery
1X-Tech · 2Xtech-PandaAI-Waton Joint Lab · 3Monash University +2
AlphaRLM: Aligning Language Models for Literature-Grounded Symbolic Factor Discovery
Oct 10, 2026Jie Huang, Yuecheng Hong, Yao Chen +6Quantitative Finance Algorithmic Trading
Harbin Institute of Technology · Independent Researcher · University of Science and Technology of China

May 14, 2026J/K move · Enter open · S save

From Feedback Loops to Policy Updates: Reinforcement Fine-Tuning for LLM-Based Alpha Factor Discovery

Lingzhe Zhang, Tong Jia, Yunpeng Zhai, Zixuan Xie+4

Peking University, Beijing, China · Alibaba Group, China · Nanjing University, Nanjing, China+1

RL for Language Models Alpha Factor Discovery Large Language Model-Guided Optimization RL Fine-Tuning Algorithmic Trading

Open paper AskPDF arXiv

Modern quantitative trading increasingly relies on systematic models to extract predictive signals from large-scale financial data, where alpha factor discovery plays a central role in transforming market observations into tradable signals. Recent LLM-based methods have shown promise in automating factor generation, but most of them still rely on prompt-level generation--evaluation--feedback loops for iterative optimization. As the loop becomes longer, repeatedly appended historical candidates and feedback can cause context explosion, increase inference cost, dilute useful information, and introduce feedback drift. Moreover, these methods often depend on very large LLMs whose stable generation preferences may lead to structurally similar expressions, redundant candidates, and search stagnation. To address these limitations, we propose QuantEvolver, a self-evolving alpha factor discovery framework based on reinforcement fine-tuning. Instead of accumulating feedback in the prompt, QuantEvolver converts executable quantitative evaluation into policy updates, enabling a Miner LLM to internalize historical optimization experience through parameter learning. Specifically, QuantEvolver constructs high-quality seed factors, builds diverse seed--time-window training tasks, generates executable Factor DSL expressions, evaluates them through Regime Backtest, and optimizes the Miner LLM with Diversity-Complementarity Reward. During training, high-quality factors are continuously accumulated in a Mined Factor Database, which serves as the final discovered factor library. Extensive experiments on three realistic market benchmarks demonstrate the effectiveness of QuantEvolver, which consistently improves the primary evaluation metric of each task over existing LLM-based alpha factor discovery baselines, produces higher-quality and more complementary factor pools.

Abstract

Alpha Factor Discovery

Explore similar work

From Feedback Loops to Policy Updates: Reinforcement Fine-Tuning for LLM-Based Alpha Factor Discovery

AlphaSchema: Exploring the Space of Trading Semantics for LLM-Based Alpha Mining

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