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

Korean

자발적으로 진화하는 배터리 에이전트는 연구 피드백을 재사용 가능한 기술, 도구 및 연구 규칙으로 전환하는 것을 목표로합니다. 이러한 축적 된 능력은 나중에 연구를 개선하도록 계속하는지 여부는 통제 평가가 필요합니다. Long-horizon alpha discovery는 국가에 의존하는 설정을 제공합니다 : 새로운 요소가 포트폴리오에 들어가면 예측 정보가 이미 커버되므로 동일한 후보자 또는 경험에 대한 가치가 시간이 지남에 따라 변할 수 있습니다. 우리는 EverMine을 도입합니다. 고정 또는 진화하는 Cap과 완전한 실험적 프레임 워크를 소개하고, 긍정적 인 경계 알파 발견에서 자발적인 경계 연구 능력을 연구합니다. EverMine은 역사의 상태 (Hist)로 연구

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