{"id":"https://openalex.org/W7160869293","doi":"https://doi.org/10.48550/arxiv.2605.08083","title":"LLMs Improving LLMs: Agentic Discovery for Test-Time Scaling","display_name":"LLMs Improving LLMs: Agentic Discovery for Test-Time Scaling","publication_year":2026,"publication_date":"2026-05-08","ids":{"openalex":"https://openalex.org/W7160869293","doi":"https://doi.org/10.48550/arxiv.2605.08083"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.08083","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08083","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.08083","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135873872","display_name":"Tong Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Tong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135909281","display_name":"Haolin Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Haolin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135845353","display_name":"Chengsong Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Chengsong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5065208258","display_name":"Huiwen Bao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bao, Huiwen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135843008","display_name":"Sheng Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Sheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135873939","display_name":"Rui Liu","orcid":"https://orcid.org/0000-0002-8227-6064"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Rui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135909516","display_name":"Runpeng Dai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dai, Runpeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135903196","display_name":"Ruibo Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Ruibo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135900971","display_name":"Chenxi Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Chenxi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135857306","display_name":"Tianyi Xiong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiong, Tianyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135881060","display_name":"Xidong Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Xidong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135844824","display_name":"Hongming Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Hongming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135882943","display_name":"Heng Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Heng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"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/T10028","display_name":"Topic Modeling","score":0.3928999900817871,"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/T10028","display_name":"Topic Modeling","score":0.3928999900817871,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.328000009059906,"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/T12072","display_name":"Machine Learning and Algorithms","score":0.047200001776218414,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/heuristics","display_name":"Heuristics","score":0.8544999957084656},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5806999802589417},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.489300012588501},{"id":"https://openalex.org/keywords/trace","display_name":"TRACE (psycholinguistics)","score":0.4699000120162964},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.4489000141620636},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.4212999939918518}],"concepts":[{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.8544999957084656},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6991999745368958},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5806999802589417},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.489300012588501},{"id":"https://openalex.org/C75291252","wikidata":"https://www.wikidata.org/wiki/Q1315756","display_name":"TRACE (psycholinguistics)","level":2,"score":0.4699000120162964},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.4489000141620636},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.4212999939918518},{"id":"https://openalex.org/C203479927","wikidata":"https://www.wikidata.org/wiki/Q5165939","display_name":"Controller (irrigation)","level":2,"score":0.4023999869823456},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3783999979496002},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.35899999737739563},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3386000096797943},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.2915000021457672},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2809999883174896},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.2793000042438507},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.2639000117778778}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.08083","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08083","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.08083","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.08083","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.8119465708732605,"display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Test-time":[0],"scaling":[1],"(TTS)":[2],"has":[3],"become":[4],"an":[5,46],"effective":[6],"approach":[7],"for":[8,90],"improving":[9],"large":[10],"language":[11],"model":[12,188],"performance":[13],"by":[14,35,150],"allocating":[15],"additional":[16],"computation":[17],"during":[18],"inference.":[19],"However,":[20],"existing":[21],"TTS":[22,57,62,91,100,157],"strategies":[23,63,169,182],"are":[24],"largely":[25],"hand-crafted:":[26],"researchers":[27,53],"manually":[28,177],"design":[29],"reasoning":[30,106,163],"patterns":[31],"and":[32,85,108,122,141,187,197,202],"tune":[33],"heuristics":[34,58],"intuition,":[36],"leaving":[37],"much":[38],"of":[39],"the":[40,76,81,138,152,167,171,191],"computation-allocation":[41],"space":[42,83],"unexplored.":[43],"We":[44,131],"propose":[45],"environment-driven":[47],"framework,":[48],"AutoTTS,":[49],"that":[50,166],"changes":[51],"what":[52],"design:":[54],"from":[55],"individual":[56],"to":[59,70,115,136,146,184],"environments":[60],"where":[61,111],"can":[64,123],"be":[65,124,205],"discovered":[66,168,181],"automatically.":[67],"The":[68,180],"key":[69],"AutoTTS":[71],"lies":[72],"in":[73],"environment":[74,78],"construction:":[75],"discovery":[77,148,193],"must":[79],"make":[80,137],"control":[82],"tractable":[84,140],"provide":[86],"cheap,":[87],"frequent":[88],"feedback":[89,145],"search.":[92],"As":[93],"a":[94,156],"concrete":[95],"instantiation,":[96],"we":[97],"formulate":[98],"width--depth":[99],"as":[101],"controller":[102],"synthesis":[103],"over":[104,175],"pre-collected":[105],"trajectories":[107],"probe":[109],"signals,":[110],"controllers":[112],"decide":[113],"when":[114],"branch,":[116],"continue,":[117],"probe,":[118],"prune,":[119],"or":[120],"stop":[121],"evaluated":[125],"cheaply":[126],"without":[127],"repeated":[128],"LLM":[129],"calls.":[130],"further":[132],"introduce":[133],"beta":[134],"parameterization":[135],"search":[139],"fine-grained":[142],"execution":[143],"trace":[144],"improve":[147,170],"efficiency":[149],"helping":[151],"agent":[153],"diagnose":[154],"why":[155],"program":[158],"fails.":[159],"Experiments":[160],"on":[161],"mathematical":[162],"benchmarks":[164,186],"show":[165],"overall":[172],"accuracy--cost":[173],"tradeoff":[174],"strong":[176],"designed":[178],"baselines.":[179],"generalize":[183],"held-out":[185],"scales,":[189],"while":[190],"entire":[192],"costs":[194],"only":[195],"$39.9":[196],"160":[198],"minutes.":[199],"Our":[200],"data,":[201],"code":[203],"will":[204],"open-source":[206],"at":[207],"https://github.com/zhengkid/AutoTTS.":[208]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-12T00:00:00"}