{"id":"https://openalex.org/W7138932284","doi":"https://doi.org/10.48550/arxiv.2603.17919","title":"Training Diffusion Language Models for Black-Box Optimization","display_name":"Training Diffusion Language Models for Black-Box Optimization","publication_year":2026,"publication_date":"2026-03-18","ids":{"openalex":"https://openalex.org/W7138932284","doi":"https://doi.org/10.48550/arxiv.2603.17919"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.17919","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17919","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.2603.17919","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5039756965","display_name":"Zipeng Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Zipeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129867770","display_name":"Can Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Can","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129861367","display_name":"Ye Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan, Ye","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129914570","display_name":"Haolun Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Haolun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120731730","display_name":"Jiayao Gu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gu, Jiayao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130161175","display_name":"Christopher Pal","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pal, Christopher","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130155173","display_name":"Xue Nan Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Xue","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/T11948","display_name":"Machine Learning in Materials Science","score":0.5539000034332275,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11948","display_name":"Machine Learning in Materials Science","score":0.5539000034332275,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10848","display_name":"Advanced Multi-Objective Optimization Algorithms","score":0.09459999948740005,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.04190000146627426,"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/leverage","display_name":"Leverage (statistics)","score":0.7245000004768372},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.5260000228881836},{"id":"https://openalex.org/keywords/domain-adaptation","display_name":"Domain adaptation","score":0.4984000027179718},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.4262000024318695},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.41620001196861267},{"id":"https://openalex.org/keywords/bridge","display_name":"Bridge (graph theory)","score":0.41609999537467957}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7681000232696533},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.7245000004768372},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5708000063896179},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.5260000228881836},{"id":"https://openalex.org/C2776434776","wikidata":"https://www.wikidata.org/wiki/Q19246213","display_name":"Domain adaptation","level":3,"score":0.4984000027179718},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4763000011444092},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.4262000024318695},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.41620001196861267},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.41609999537467957},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.399399995803833},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.3806999921798706},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.3749000132083893},{"id":"https://openalex.org/C92548554","wikidata":"https://www.wikidata.org/wiki/Q2262868","display_name":"Domain model","level":3,"score":0.36090001463890076},{"id":"https://openalex.org/C88006597","wikidata":"https://www.wikidata.org/wiki/Q690117","display_name":"Disk formatting","level":2,"score":0.3075000047683716},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.29100000858306885},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2680000066757202}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.17919","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17919","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.2603.17919","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17919","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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,121],"study":[1],"offline":[2,13,68],"black-box":[3],"optimization":[4],"(BBO),":[5],"aiming":[6],"to":[7,36,49,67,70,113,128],"discover":[8],"improved":[9],"designs":[10,16],"from":[11],"an":[12],"dataset":[14,146],"of":[15,86],"and":[17,24,89,97,109,150],"labels,":[18],"a":[19,77,105,124],"problem":[20],"common":[21],"in":[22,56,93],"robotics":[23],"DNA":[25],"with":[26,134,157,173],"limited":[27],"labeled":[28],"samples.":[29],"While":[30],"recent":[31],"work":[32,189],"applies":[33],"autoregressive":[34],"LLMs":[35,66,88],"BBO":[37,69,94],"by":[38,160],"formatting":[39],"tasks":[40],"as":[41],"natural-language":[42],"prompts,":[43],"their":[44,72],"left-to-right":[45],"design":[46,57],"generation":[47,133],"struggles":[48],"capture":[50],"the":[51,82,90,130,144,151],"strong":[52],"bidirectional":[53,73],"dependencies":[54],"inherent":[55],"problems.":[58],"To":[59,99],"address":[60],"this,":[61],"we":[62,103],"propose":[63,123],"adapting":[64],"diffusion":[65,87,131],"leverage":[71],"modeling":[74],"capabilities.":[75],"However,":[76],"domain":[78,119],"gap":[79],"exists":[80],"between":[81],"natural":[83],"text":[84],"pre-training":[85],"heterogeneous":[91],"signals":[92],"(prompts,":[95],"designs,":[96],"labels).":[98],"bridge":[100],"this":[101],"gap,":[102],"construct":[104],"unified":[106,145],"prompt--response":[107],"corpus":[108],"introduce":[110],"delimiter":[111],"tokens":[112],"explicitly":[114],"mark":[115],"field":[116],"boundaries":[117],"for":[118,183,187],"adaptation.":[120],"further":[122],"two-stage":[125],"post-training":[126],"framework":[127],"align":[129],"LLM":[132],"high-label":[135],"designs.":[136],"The":[137],"first":[138],"stage":[139,153],"performs":[140],"supervised":[141],"fine-tuning":[142],"on":[143,168],"via":[147],"masked-response":[148],"prediction,":[149],"second":[152],"adopts":[154],"reinforcement":[155],"learning":[156],"rewards":[158],"defined":[159],"label":[161],"improvements.":[162],"Our":[163],"method":[164],"achieves":[165],"state-of-the-art":[166],"results":[167],"Design-Bench":[169],"under":[170],"small-data":[171],"settings":[172],"highly":[174],"efficient":[175],"training,":[176],"requiring":[177],"only":[178],"$1.5$":[179],"H100":[180],"GPU":[181],"hours":[182],"discrete":[184],"tasks.":[185],"Code":[186],"our":[188],"is":[190],"available":[191],"here:":[192],"https://github.com/zpointS/DiBO.":[193]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-03-20T00:00:00"}