{"id":"https://openalex.org/W7213580804","doi":"https://doi.org/10.48550/arxiv.2609.19526","title":"Self Improvement via Fast Tree-search","display_name":"Self Improvement via Fast Tree-search","publication_year":2026,"publication_date":"2026-09-17","ids":{"openalex":"https://openalex.org/W7213580804","doi":"https://doi.org/10.48550/arxiv.2609.19526"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.19526","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.19526","pdf_url":"https://arxiv.org/pdf/2609.19526","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.19526","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5152140870","display_name":"Xinghong Fu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Fu, Xinghong","raw_affiliation_strings":["Massachusetts Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029725024","display_name":"Aravinth Kulanthaivelu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Kulanthaivelu, Aravinth","raw_affiliation_strings":["Massachusetts Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology","institution_ids":["https://openalex.org/I63966007"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5152120627","display_name":"Yutaro Yamada","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I63966007","display_name":"Massachusetts Institute of Technology","ror":"https://ror.org/042nb2s44","country_code":"US","type":"education","lineage":["https://openalex.org/I63966007"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yamada, Yutaro","raw_affiliation_strings":["Massachusetts Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Massachusetts Institute of Technology","institution_ids":["https://openalex.org/I63966007"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I63966007"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":7,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11975","display_name":"Evolutionary Algorithms and Applications","score":0.5946999788284302,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11975","display_name":"Evolutionary Algorithms and Applications","score":0.5946999788284302,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10260","display_name":"Software Engineering Research","score":0.37599998712539673,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10533","display_name":"Teaching and Learning Programming","score":0.00800000037997961,"subfield":{"id":"https://openalex.org/subfields/1706","display_name":"Computer Science Applications"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/personal-development","display_name":"human development","score":0.7580000162124634},{"id":"https://openalex.org/keywords/ai-coding-agents","display_name":"coding agents","score":0.6790000200271606},{"id":"https://openalex.org/keywords/tree-search","display_name":"tree search","score":0.656000018119812},{"id":"https://openalex.org/keywords/llm-as-a-judge","display_name":"LLM-as-a-judge","score":0.8379999995231628},{"id":"https://openalex.org/keywords/recursive-self-modification","display_name":"recursive self-modification","score":0.6269999742507935},{"id":"https://openalex.org/keywords/sample-efficiency","display_name":"sample efficiency","score":0.5870000123977661},{"id":"https://openalex.org/keywords/bradley-terry-model","display_name":"Bradley-Terry model","score":0.7820000052452087},{"id":"https://openalex.org/keywords/pairwise-comparison","display_name":"pairwise comparison","score":0.7480000257492065},{"id":"https://openalex.org/keywords/code-generation","display_name":"code generation","score":0.37599998712539673}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7444999814033508},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.6385999917984009},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5735999941825867},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.5033000111579895},{"id":"https://openalex.org/C113174947","wikidata":"https://www.wikidata.org/wiki/Q2859736","display_name":"Tree (set theory)","level":2,"score":0.4706000089645386},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.4546999931335449},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.4113999903202057},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.3873000144958496},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.37299999594688416},{"id":"https://openalex.org/C163797641","wikidata":"https://www.wikidata.org/wiki/Q2067937","display_name":"Tree structure","level":3,"score":0.3718000054359436},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3582000136375427},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.34450000524520874},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.3231000006198883},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.30169999599456787},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3003000020980835},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.28940001130104065},{"id":"https://openalex.org/C180591934","wikidata":"https://www.wikidata.org/wiki/Q1253369","display_name":"Downtime","level":2,"score":0.2831999957561493},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.25200000405311584}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.19526","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.19526","pdf_url":"https://arxiv.org/pdf/2609.19526","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.19526","doi":"https://doi.org/10.48550/arxiv.2609.19526","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.19526","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.19526","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.19526","pdf_url":"https://arxiv.org/pdf/2609.19526","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7213580804.pdf","grobid_xml":"https://content.openalex.org/works/W7213580804.grobid-xml"},"referenced_works_count":12,"referenced_works":["https://openalex.org/W2155277832","https://openalex.org/W2739848020","https://openalex.org/W2951365061","https://openalex.org/W3083410900","https://openalex.org/W4388926373","https://openalex.org/W4402500863","https://openalex.org/W4407247211","https://openalex.org/W4412505633","https://openalex.org/W4412889810","https://openalex.org/W4414939355","https://openalex.org/W7127903492","https://openalex.org/W7166060389"],"related_works":["https://openalex.org/W7214890487","https://openalex.org/W7223573068","https://openalex.org/W7101430733","https://openalex.org/W7223215301","https://openalex.org/W7221813947","https://openalex.org/W7202216379","https://openalex.org/W7221768944","https://openalex.org/W7208811582","https://openalex.org/W7214127912","https://openalex.org/W7166855075"],"abstract_inverted_index":{"Coding":[0],"agents":[1],"can":[2,18],"recursively":[3],"modify":[4],"their":[5,61],"own":[6],"implementations,":[7],"forming":[8],"a":[9,32,65,110,124,142],"loop":[10],"of":[11,49,67,187],"self-improvement.":[12],"While":[13],"prior":[14,58],"work":[15],"shows":[16],"this":[17],"boost":[19],"performance":[20,41],"on":[21,157,175],"coding":[22,40],"benchmarks,":[23],"existing":[24,170],"approaches":[25,59],"are":[26,133],"costly":[27],"and":[28,114,193],"compute-intensive.":[29],"We":[30,46,77],"introduce":[31,78],"simple,":[33],"sample-efficient":[34],"self-improvement":[35],"framework":[36],"that":[37,96],"significantly":[38,181],"improves":[39],"under":[42],"strict":[43],"budget":[44],"constraints.":[45],"identify":[47],"evaluation":[48,166],"candidate":[50,101,159],"self-modifications":[51],"as":[52],"the":[53,71,104,115,137,148,176],"main":[54],"runtime":[55],"bottleneck":[56],"since":[57],"estimate":[60],"effectiveness":[62],"by":[63,164],"re-running":[64],"subset":[66],"benchmark":[68,179],"tasks":[69],"with":[70,92,109,180],"modified":[72],"agent,":[73],"which":[74,86],"is":[75,107],"time-consuming.":[76],"Recursive":[79],"Self":[80],"Improvement":[81],"via":[82],"Fast":[83],"Tree-search":[84],"(SIFT),":[85],"augments":[87],"these":[88],"downstream":[89,130],"task":[90,131],"evaluations":[91,132],"an":[93],"LLM-as-a-judge":[94],"signal":[95,153],"performs":[97],"pairwise":[98],"comparisons":[99],"between":[100],"patches,":[102],"where":[103],"win-loss":[105],"record":[106],"aggregated":[108],"regularized":[111],"Bradley-Terry":[112],"model,":[113],"resulting":[116],"strength":[117],"scores":[118,150],"drive":[119],"rank-based":[120],"parent":[121],"sampling":[122],"inside":[123],"lightweight":[125],"disaggregated":[126,144],"tree":[127,145],"search.":[128],"Expensive":[129],"reserved":[134],"only":[135],"for":[136],"most":[138],"promising":[139,158],"nodes.":[140],"Using":[141],"fully":[143],"search":[146],"pipeline,":[147],"judge":[149],"provide":[151],"intermediate":[152],"to":[154],"guide":[155],"exploration":[156],"patches":[160],"without":[161],"being":[162],"bottlenecked":[163],"slow":[165],"runs.":[167],"SIFT":[168],"outperforms":[169],"tree-search":[171],"based":[172],"self-evolution":[173],"frameworks":[174],"full":[177],"Polyglot":[178],"lower":[182],"resource":[183],"requirements":[184],"in":[185],"terms":[186],"CPU":[188],"hours,":[189],"wall":[190],"clock":[191],"time,":[192],"API":[194],"cost.":[195]},"counts_by_year":[{"year":2026,"cited_by_count":7}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-09-19T00:00:00"}