[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85115-en":3,"doc-seo-85115-105":29,"detail-sidebar-cat-0-en-105":91},{"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":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},85115,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Ideas Have Genomes Benchmarking Scientific Lineage Reasoning and Lineage-Grounded Idea Generation","Scientific ideas evolve from prior work rather than appearing from a blank page, combining inheritance, repair of known limitations, and recombination of mechanisms. The document introduces IdeaGeneBench (IG-Bench), a benchmark for scientific lineage reasoning and lineage-grounded idea generation using an IdeaGene framework with typed, evidence-grounded Idea Genome objects aligned via GenomeDiff. IG-Bench includes golden lineage traces across 10 domains and supports IG-Exam for closed-form lineage reasoning and IG-Arena for lineage-conditioned proposal generation.","arXiv :2607 .08758v 1 [ cs .AI] 9 Jul 2026  \nIdeas Have Genomes: Benchmarking Scientific Lineage Reasoning  \nand Lineage-Grounded Idea Generation  \nYifan Zhou 1 ,∗ , Qihao Yang 1 ,∗ , Yan Li 1 ,∗ , Donggang Li 1 ,∗ , Xiru Hu 1 , Hokin Deng2 , Ziyang Gong 1 ,  \nXuanyi Zhou 1 , Huacan Wang3 , Xiangchao Yan4 , Wanghan Xu 1 , Wenlong Zhang4 , Shaofeng Zhang5 , Yue Zhou6 , Yifan Yang7 , Zhihang Zhong 1 , Xue Yang 1 ,†  \n1 Shanghai Jiao Tong University, 2 Carnegie Mellon University, 3 University of Chinese Academy of Sciences, 4 Shanghai Artificial Intelligence Laboratory, 5 University of Science and Technology of China,  \n6 East China Normal University, 7 Microsoft Corporation  \n∗ Equal contribution, †Corresponding Author  \nAbstract  \nScientific ideas rarely start from a blank page. They inherit mechanisms, repair known limitations, and recombine pieces of earlier work, much like biological genomes. Current benchmarks still say little about whether AI systems can follow this inheritance structure. We present IdeaGeneBench (IG-Bench), a benchmark for scientific lineage reasoning and lineage-grounded idea generation. IG-Bench is organized around the IdeaGene framework: each paper or proposal is represented as a set of minimal, typed, evidence-grounded Idea Genome objects, and a GenomeDiff aligns these objects to record inheritance, mutation, loss, external import, and novel insertion under six operational evolutionary dynamics. The benchmark contains 1,961 golden lineage traces, 1,085 curated Idea Genome objects, and 920 pairwise GenomeDiff records across 10 scientific domains. It supports two evaluations. IG-Exam (42 task types, 1,029 instances) tests closed-form lineage reasoning across Idea Genome abstraction, inheritance tracing, evolutionary reasoning, and lineage verification. IG-Arena evaluates generation with a lineage-conditioned Population-Evolution Score (PES), asking whether a proposal can be inserted as a coherent descendant of a given lineage population: it should inherit the right Idea Genome objects, vary meaningfully from nearby work, and offer selection value for future research. Experiments on 14 LLM-based scientists expose a compositional bottleneck. The strongest system reaches only 27 .3% exact accuracy on lineage reasoning, and structured lineage context reshuffles system rankings rather than helping every participant uniformly.  \nDate: July 10, 2026  \nCorrespondence: [yangxue-2019-sjtu@sjtu.edu.cn](yangxue-2019-sjtu@sjtu.edu.cn)  \nProject Page: [https://visionxlab.github.io/IdeasHaveGenomes/](https://visionxlab.github.io/IdeasHaveGenomes/)  \n“Descent with modification.”  \n— Charles Darwin, On the Origin of Species (1859)  \nFigure 1 From paper-centric search to genome-centric lineage. A paper-centric view leaves a model with many related papers but no explicit unit for deciding which ideas are inherited, mutated, recombined, or merely co-located. The genome-centric view first extracts Idea Genome objects, then aligns them through GenomeDiff records into populations and lineages, making evolutionary relations explicit enough to evaluate.  \n1 Introduction  \nLLM-based auto-research systems now search literature, synthesize hypotheses, run experiments, and write paper-like reports [25, 40 , 50] . Existing evaluations emphasize retrieval quality, factuality, writing fluency, novelty, workflow competence, or pairwise preference [1, 37 , 39 , 44 , 45 , 48 , 53] . These criteria leave a harder question mostly untested: when a proposal claims to extend a research direction, does it inherit the right mechanism, repair the right limitation, and stay coherent with the lineage it builds on? Figure 1 motivates this shift from paper-centric context gathering to genome-centric lineage comparison.  \nThe distinction matters because scientific progress is not the same as topical proximity. Two papers can share a task without inheriting from one another, and two papers can look textually distant while carrying the same core mechan","cbCaicX13oiz97GS","https://ap.wps.com/l/cbCaicX13oiz97GS","pdf",3803570,1,22,"English","en",105,"# Abstract\n# Introduction\n## Limitations of existing benchmarks\n## Scientific lineage competence\n## IdeaGene and evaluable lineage objects\n## IdeaGene-Bench (IG-Bench) overview","[{\"question\":\"What problem does IG-Bench address in evaluating AI scientific auto-research systems?\",\"answer\":\"IG-Bench targets the gap where existing evaluations rarely test whether a proposal truly inherits the correct mechanisms and repairs the right limitations within a coherent research lineage, rather than only matching topical proximity.\"},{\"question\":\"How does the IdeaGene framework represent scientific papers for lineage comparison?\",\"answer\":\"Each paper or proposal is represented as a set of minimal, typed, evidence-grounded Idea Genome objects, enabling explicit auditing of idea inheritance, mutation, loss, and insertion across work.\"},{\"question\":\"What are the two evaluation settings in IG-Bench?\",\"answer\":\"IG-Exam evaluates closed-form lineage reasoning across Idea Genome abstraction, inheritance tracing, evolutionary reasoning, and lineage verification, while IG-Arena assesses generation under a lineage-conditioned Population-Evolution Score (PES).\"}]",1784201191,55,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"ideas-have-genomes-benchmarking-scientific-lineage-reasoning-and-lineage-grounded-idea-generation","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/ideas-have-genomes-benchmarking-scientific-lineage-reasoning-and-lineage-grounded-idea-generation/85115/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does IG-Bench address in evaluating AI scientific auto-research systems?","Question",{"text":75,"@type":76},"IG-Bench targets the gap where existing evaluations rarely test whether a proposal truly inherits the correct mechanisms and repairs the right limitations within a coherent research lineage, rather than only matching topical proximity.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the IdeaGene framework represent scientific papers for lineage comparison?",{"text":80,"@type":76},"Each paper or proposal is represented as a set of minimal, typed, evidence-grounded Idea Genome objects, enabling explicit auditing of idea inheritance, mutation, loss, and insertion across work.",{"name":82,"@type":73,"acceptedAnswer":83},"What are the two evaluation settings in IG-Bench?",{"text":84,"@type":76},"IG-Exam evaluates closed-form lineage reasoning across Idea Genome abstraction, inheritance tracing, evolutionary reasoning, and lineage verification, while IG-Arena assesses generation under a lineage-conditioned Population-Evolution Score (PES).","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]