Newest papers

EverMine: Dissecting the Self-Evolution of Research Capabilities in Long-Horizon Alpha Research

Siyuan Li, Jiangfeng Zhang, Rui Yao, Weihua Qiu, Mingyang Xu, Zixuan Yuan · 2026 · arXiv (Cornell University)

DOI: 10.48550/arxiv.2609.33524 · cited by 0 works (OpenAlex) · open-access PDF

French

Les agents auto-évolutifs de bat ont pour objectif de transformer le feedback de la recherche en compétences, outils et règles de recherche réutilisables. Que ces capacités accumulées continuent d'améliorer la recherche ultérieure nécessite une évaluation contrôlée. La découverte alpha à long horizon fournit un environnement dépendant de l'état: une fois qu'un nouveau facteur entre dans le portefeuille, les informations prédictives sont déjà couvertes, de sorte que la valeur du même candidat ou de l'expérience peut changer au fil du temps. Nous introduisons EverMine, un cadre empirique pour étudier les capacités de recherche en auto-évolution dans la découverte alpha à long horizon. EverMine décompose la recherche dans l'état (Hist), le portefeuille

Machine translation (Meta M2M100); the English original follows.

Abstract

Self-evolving agents aim to turn research feedback into reusable skills, tools, and research rules. Whether these accumulated capabilities continue to improve later research requires controlled evaluation. Long-horizon alpha discovery provides a state-dependent setting: once a new factor enters the portfolio, the predictive information already covered changes, so the value of the same candidate or experience may change over time. We introduce EverMine, an empirical framework for studying self-evolving research capabilities in long-horizon alpha discovery. EverMine decomposes the research state into history (Hist), the current factor portfolio (Frontier), and reusable capabilities (Cap). Under matched resource limits, we compare complete runs with fixed or evolving Cap, and replace Cap while holding Hist and Frontier fixed to estimate the conditional value of accumulated capabilities. We also combine full trajectories with historical-state replay to examine how experience-based decisions affect candidate selection and portfolio outcomes. Across 18 long-horizon trajectories, end-to-end comparisons show no consistent gain from Cap evolution. Across 48 continuation branches from shared Hist and Frontier states, accumulated Cap also does not consistently outperform the initial Cap. Parameter tuning of existing factor structures can still improve the portfolio. In an exploratory replay of two screening batches from one Evolving trajectory, some screened-out candidates have positive marginal value at the original state, yet submitting all screened-out candidates sequentially slightly lowers final portfolio IC in both batches. These results show that candidate value depends on the evolving portfolio and submission order, and motivate evaluating self-evolving research capabilities through end-to-end outcomes, conditional capability value, and the consequences of experience-based decisions.

Read in: English · Spanish · Chinese · Hindi · Arabic · Portuguese · French · German · Japanese · Russian · Korean · Indonesian · Vietnamese

Topics: Portfolio, Computer science, Efficient frontier, Resource (disambiguation), Value (mathematics), Selection (genetic algorithm)

Sources: OpenAlex (CC0), arXiv, Crossref. Agent Exchange