[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84356-en":3,"doc-seo-84356-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},84356,1099514068365,"Aurelia","https://ap-avatar.wpscdn.com/avatar/10000253d8d9f28188e?_k=1776742907772140068",8,"Research & Report","XALPHA: Memory-Driven AI Quant Researcher for Hypothesis-to-Code Alpha Discovery","Financial markets are noisy, non-stationary, and high-dimensional, making robust predictive trading-signal discovery difficult. Alpha discovery has progressed from manual factor design to ML, evolutionary search, and LLM-based frameworks, yet most approaches automate isolated steps rather than running an end-to-end hypothesis-to-code workflow. XALPHA introduces a memory-driven AI quant researcher with multi-source research memory and macro/micro/cross brain roles to plan themes, generate executable factor code, validate ex-ante tri-alignment, and consolidate feedback. Experiments on CSI300 show stronger overall performance than baselines.","arXiv :2607 .08332v2 [ cs .CL] 13 Jul 2026  \nXALPHA: A Memory-Driven AI Quant Researcher for  \nHypothesis-to-Code Alpha Discovery  \nFengyuan Liu 1 ,2 Yuchen Fu 1 Yuqi Wang 1 Qi Liu 1 ,2  \n1 School of Computing and Data Science, The University of Hong Kong  \n2 Grace Investment Machine  \n[oxfengyuan@gmail.com](oxfengyuan@gmail.com)  \nAbstract  \nFinancial markets are noisy, non-stationary, and high-dimensional, making it difficult to discover predictive and robust trading signals. Alpha discovery has evolved from manual factor design to machine learning, evolutionary search, and recent LLM-based frameworks, improving the efficiency of factor generation, search, and evaluation. However, existing methods still mostly automate isolated steps, rather than functioning as end-to-end quant researchers that can absorb external knowledge, close the hypothesis-to-code validation loop, and learn from accumulated discovery feedback. To fill this gap, we introduce XALPHA, a memory-driven AI Quant Researcher for continuous hypothesis-to-code alpha discovery. XALPHA maintains a multi-source research memory system that integrates report-grounded financial knowledge with discovery feedback from prior generations and research cycles. Guided by this memory system, a Macro Brain plans research themesand selects suitable Archetypes; a Micro Brain transforms the planned hypothesis pool into executable factor code and verifies ex-ante tri-alignment among the hypothesis idea, code logic, and financial plausibility; and a Cross Brain consolidates empirical outcomes into generation-level feedback, cycle-level summaries, and archetype-level research cues for future exploration. In this way, XALPHA turns alpha mining from isolated factor generation into a closed-loop research process that continuously reads, hypothesizes, implements, validates, reflects, and evolves. Experiments on CSI300 show that XALPHA achieves stronger overall alpha discovery performance than representative baselines.  \n1 Introduction  \nAlpha mining aims to discover predictive trading signals, or alphas, that map historical market observations into forecasts of future asset returns. It is a central problem in quantitative investing, yet remains difficult because financial markets are noisy, non-stationary, and characterized by timevarying volatility [Engle, 1982] . This difficulty has driven alpha discovery from manual factor design toward increasingly automated, explainable, and knowledge-driven approaches [Guo et al., 2024] . However, current automation still mostly addresses isolated factor-level operations, such as searching, generating, or evaluating candidate factors, rather than the full hypothesis-driven workflow of a human quant researcher. In practice, alpha discovery is a hypothesis-to-code loop: researchers absorb market knowledge, formulate financial hypotheses, translate them into factor implementations, examine their financial plausibility, evaluate them empirically, and reuse both successes and failures to guide future exploration. This gap motivates our goal of building an AI Quant Researcher that conducts alpha discovery as an iterative, memory-driven research process, rather than as isolated factor generation.  \nPreprint.  \nTraditional alpha discovery largely relies on human expertise. Researchers manually design factors based on economic intuition, empirical asset-pricing evidence, and market observations [Fama and French, 1992, 1993, Kakushadze, 2016] . These human-crafted signals are often interpretable and easy to audit, but the discovery process is labor-intensive, difficult to scale, and heavily dependent on individual researchers’ accumulated experience. As markets become more complex and financial datasets, reports, and research materials continue to proliferate, manual factor design struggles to support continuous hypothesis exploration and systematic knowledge accumulation.  \nTo improve the scalability of alpha discovery, machine learning and automated searc","cbCaitn9kpJxQVzO","https://ap.wps.com/l/cbCaitn9kpJxQVzO","pdf",2323107,7,1,61,"English","en",105,"# Abstract\n# Introduction\n## Problem of Alpha Mining\n## Traditional Human-Curated Discovery\n## Machine Learning and Automated Search Approaches\n## LLM-Based Alpha Mining and Multi-Agent Workflows","[{\"question\":\"What problem does XALPHA address in alpha discovery?\",\"answer\":\"XALPHA addresses the gap where existing methods automate isolated factor steps instead of executing a full hypothesis-to-code loop that absorbs knowledge, validates, and learns from accumulated feedback.\"},{\"question\":\"How does XALPHA use memory to improve research?\",\"answer\":\"XALPHA maintains a multi-source research memory that integrates report-grounded financial knowledge with discovery feedback from previous generations and research cycles.\"},{\"question\":\"What are the roles of the Macro Brain, Micro Brain, and Cross Brain?\",\"answer\":\"The Macro Brain plans research themes and selects archetypes, the Micro Brain converts hypotheses into executable factor code and checks ex-ante tri-alignment (idea, code logic, financial plausibility), and the Cross Brain consolidates empirical results into feedback for future cycles.\"}]",1784195047,154,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"xalpha-memory-driven-ai-quant-researcher-for-hypothesis-to-code-alpha-discovery","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/xalpha-memory-driven-ai-quant-researcher-for-hypothesis-to-code-alpha-discovery/84356/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-07-27","2026-07-16",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What problem does XALPHA address in alpha discovery?","Question",{"text":76,"@type":77},"XALPHA addresses the gap where existing methods automate isolated factor steps instead of executing a full hypothesis-to-code loop that absorbs knowledge, validates, and learns from accumulated feedback.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does XALPHA use memory to improve research?",{"text":81,"@type":77},"XALPHA maintains a multi-source research memory that integrates report-grounded financial knowledge with discovery feedback from previous generations and research cycles.",{"name":83,"@type":74,"acceptedAnswer":84},"What are the roles of the Macro Brain, Micro Brain, and Cross Brain?",{"text":85,"@type":77},"The Macro Brain plans research themes and selects archetypes, the Micro Brain converts hypotheses into executable factor code and checks ex-ante tri-alignment (idea, code logic, financial plausibility), and the Cross Brain consolidates empirical results into feedback for future cycles.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]