{"id":"https://openalex.org/W7155194767","doi":"https://doi.org/10.48550/arxiv.2604.19440","title":"What Makes an LLM a Good Optimizer? A Trajectory Analysis of LLM-Guided Evolutionary Search","display_name":"What Makes an LLM a Good Optimizer? A Trajectory Analysis of LLM-Guided Evolutionary Search","publication_year":2026,"publication_date":"2026-04-21","ids":{"openalex":"https://openalex.org/W7155194767","doi":"https://doi.org/10.48550/arxiv.2604.19440"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.19440","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.19440","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,"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.19440","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134308920","display_name":"Xinhao Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Xinhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134278861","display_name":"Xi Chen","orcid":"https://orcid.org/0009-0003-9293-8134"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Xi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134264944","display_name":"Fran\u00e7ois Portet","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Portet, Fran\u00e7ois","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5134302414","display_name":"Maxime Peyrard","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peyrard, Maxime","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/T10181","display_name":"Natural Language Processing Techniques","score":0.21690000593662262,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.21690000593662262,"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/T10028","display_name":"Topic Modeling","score":0.12729999423027039,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.1096000000834465,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/novelty","display_name":"Novelty","score":0.7885000109672546},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.5378999710083008},{"id":"https://openalex.org/keywords/evolutionary-algorithm","display_name":"Evolutionary algorithm","score":0.527400016784668},{"id":"https://openalex.org/keywords/optimization-problem","display_name":"Optimization problem","score":0.35580000281333923},{"id":"https://openalex.org/keywords/multi-objective-optimization","display_name":"Multi-objective optimization","score":0.3553999960422516},{"id":"https://openalex.org/keywords/evolutionary-computation","display_name":"Evolutionary computation","score":0.35019999742507935}],"concepts":[{"id":"https://openalex.org/C2778738651","wikidata":"https://www.wikidata.org/wiki/Q16546687","display_name":"Novelty","level":2,"score":0.7885000109672546},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6319000124931335},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.5378999710083008},{"id":"https://openalex.org/C159149176","wikidata":"https://www.wikidata.org/wiki/Q14489129","display_name":"Evolutionary algorithm","level":2,"score":0.527400016784668},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5238999724388123},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.44029998779296875},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.35580000281333923},{"id":"https://openalex.org/C68781425","wikidata":"https://www.wikidata.org/wiki/Q2052203","display_name":"Multi-objective optimization","level":2,"score":0.3553999960422516},{"id":"https://openalex.org/C105902424","wikidata":"https://www.wikidata.org/wiki/Q1197129","display_name":"Evolutionary computation","level":2,"score":0.35019999742507935},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.319599986076355},{"id":"https://openalex.org/C2987595161","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Optimization algorithm","level":2,"score":0.2705000042915344},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.260699987411499},{"id":"https://openalex.org/C199505168","wikidata":"https://www.wikidata.org/wiki/Q3267529","display_name":"Evolutionary robotics","level":3,"score":0.2581999897956848},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.2540999948978424},{"id":"https://openalex.org/C109718341","wikidata":"https://www.wikidata.org/wiki/Q1385229","display_name":"Metaheuristic","level":2,"score":0.2538999915122986},{"id":"https://openalex.org/C135320971","wikidata":"https://www.wikidata.org/wiki/Q1868524","display_name":"Local search (optimization)","level":2,"score":0.2538999915122986}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.19440","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.19440","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,"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":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.19440","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.19440","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,"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"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.519344687461853}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"work":[1],"has":[2],"demonstrated":[3],"the":[4,19,63,99,133,142,148],"promise":[5],"of":[6,36,62,120,141,150],"orchestrating":[7],"large":[8,108],"language":[9],"models":[10,65],"(LLMs)":[11],"within":[12],"evolutionary":[13,38],"and":[14,76,155,160,167],"agentic":[15],"optimization":[16,23,41,56,158],"systems.":[17],"However,":[18],"mechanisms":[20],"driving":[21],"these":[22,80],"gains":[24],"remain":[25],"poorly":[26],"understood.":[27],"In":[28],"this":[29],"work,":[30],"we":[31,82],"present":[32],"a":[33],"large-scale":[34],"study":[35],"LLM-guided":[37],"search,":[39],"collecting":[40],"trajectories":[42,75],"for":[43,153,164],"15":[44],"LLMs":[45],"across":[46],"8":[47],"tasks.":[48],"Although":[49],"zero-shot":[50],"problem-solving":[51],"ability":[52],"correlates":[53],"with":[54,66,111],"final":[55,126],"outcomes,":[57],"it":[58],"explains":[59],"only":[60,131],"part":[61],"variance:":[64],"similar":[67],"initial":[68],"capability":[69],"often":[70],"induce":[71],"dramatically":[72],"different":[73],"search":[74,100,134],"outcomes.":[77],"By":[78],"analyzing":[79],"trajectories,":[81],"find":[83],"that":[84],"strong":[85],"LLM":[86],"optimizers":[87,106],"behave":[88],"as":[89],"local":[90],"refiners,":[91],"producing":[92],"frequent":[93],"incremental":[94],"improvements":[95],"while":[96],"progressively":[97],"localizing":[98],"in":[101],"semantic":[102,109],"space.":[103,144],"Conversely,":[104],"weaker":[105],"exhibit":[107],"drift,":[110],"sporadic":[112],"breakthroughs":[113],"followed":[114],"by":[115],"stagnation.":[116],"Notably,":[117],"various":[118],"measures":[119],"solution":[121,143],"novelty":[122,128],"do":[123],"not":[124],"predict":[125],"performance;":[127],"is":[129],"beneficial":[130],"when":[132],"remains":[135],"sufficiently":[136],"localized":[137],"around":[138],"high-performing":[139],"regions":[140],"Our":[145],"results":[146],"highlight":[147],"importance":[149],"trajectory":[151],"analysis":[152],"understanding":[154],"improving":[156],"LLM-based":[157],"systems":[159],"provide":[161],"actionable":[162],"insights":[163],"their":[165],"design":[166],"training.":[168]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-04-23T00:00:00"}