arXiv:2607. 26642v1 Announce Type: new Abstract: Automated alpha mining has increasingly adopted large language model (LLM) agents for factor generation and iterative discovery.
Automated alpha mining has increasingly adopted large language model (LLM) agents for factor generation and iterative discovery. However, existing LLM-based systems often delegate both factor construction and search decisions to the agent itself, without an explicit exploration space or a principled mechanism for navigating that space.
arXiv:2603.16365v3 Announce Type: replace Abstract: We study alpha factor mining, the automated discovery of predictive signals from noisy, non-stationary market data-under a practical requirement th...
Alpha‑R1 introduces a reinforcement‑learning aligned large language model framework that performs context‑aware alpha screening by semantically gating candidate factors against a dynamic market state description. The model, trained with group relative policy optimization using realized portfolio returns as reward, selects a sparse subset of factors whose economic rationale matches current market conditions. In a 12‑month out‑of‑sample test, Alpha‑R1 achieved annualized returns of 47.87% on the S&P 500 and 40.57% on the CSI 300, with Sharpe ratios of 1.62 and 2.23, demonstrating the effectiveness of semantic factor reranking in non‑stationary markets.
arXiv:2610.06947v1 Announce Type: cross Abstract: Factor mining seeks to discover signals from financial data that predict future asset returns and guide portfolio construction. Automated factor mini...
AlphaDiverse is a framework that enhances large language model–based multi‑agent systems for alpha factor mining by addressing cost, availability, and confidentiality constraints. It generates diverse research paths through varied environments and post‑training local agents, then fine‑tunes these agents with supervised learning and optimizes them jointly using a GRPO method that balances predictive quality and diversity. The approach limits research feedback to inner‑period data and evaluates a frozen model on outer‑period data to avoid test‑set tuning, demonstrating competitive prediction and broader exploration across four Chinese stock universes.
AlphaMemo is a self‑evolving alpha‑mining agent that uses Structured Search‑Process Memory to improve factor discovery. Instead of storing only final factors or full search trajectories, it records reusable evidence about which edit motifs succeed or fail in specific parent‑factor contexts, extracted from Abstract Syntax Tree differences. The system incorporates confidence‑gated residual memory, a search‑ledger prior, and asymmetric veto control to suppress high‑confidence failure patterns, leading to better out‑of‑sample performance and fixed‑budget discovery efficiency on CSI 500 and S&P 500 datasets.
arXiv:2608. 12841v1 Announce Type: cross Abstract: We study recursive self-improvement at the level of quantitative-investment research: whether an autonomous system can use evidence from earlier experiments to improve the hypotheses and candidates proposed in later iterations.
arXiv:2608.30192v1 Announce Type: new Abstract: Traditional finance relies on experts to hand-craft factors through a principled process grounded in economic rationale. Recent LLM-based multi-agent s...
arXiv:2508. 13174v2 Announce Type: replace Abstract: Formula alpha mining, which generates predictive signals from financial data, is critical for quantitative investment.
X-Alpha is an AI Quant Researcher that uses persistent research memory and continuous hypothesis-to-code discovery to overcome key bottlenecks in automated alpha mining. It converts financial reports into actionable research paths via Report-to-Memory Abstraction, directs exploration with memory-driven routing, and refines hypotheses through executable factors and empirical validation. Experiments on five equity indices, especially CSI300, show that X-Alpha improves factor quality, signal diversity, and consistently outperforms baselines across markets and horizons.
The paper introduces LineageEvo, a method that incorporates lineage-level control into large‑language‑model (LLM) based evolutionary factor mining. It uses lineage‑conditioned LLM variation to guide semantic generation and lineage‑stratified search allocation to balance quality and diversity across evolutionary branches. Experiments on CSI 500 and S&P 500 show that LineageEvo outperforms existing approaches in predictive performance and produces a more balanced, structurally diverse factor pool.