{"id":"https://openalex.org/W7151620499","doi":"https://doi.org/10.1109/icmla66185.2025.00134","title":"A Reward-Driven Controller for Text Generation with Black-Box Language Models","display_name":"A Reward-Driven Controller for Text Generation with Black-Box Language Models","publication_year":2025,"publication_date":"2025-12-03","ids":{"openalex":"https://openalex.org/W7151620499","doi":"https://doi.org/10.1109/icmla66185.2025.00134"},"language":null,"primary_location":{"id":"doi:10.1109/icmla66185.2025.00134","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icmla66185.2025.00134","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Conference on Machine Learning and Applications (ICMLA)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5074307376","display_name":"Robert Morain","orcid":null},"institutions":[{"id":"https://openalex.org/I100005738","display_name":"Brigham Young University","ror":"https://ror.org/047rhhm47","country_code":"US","type":"education","lineage":["https://openalex.org/I100005738"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Robert Morain","raw_affiliation_strings":["Brigham Young University,Department of Computer Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Brigham Young University,Department of Computer Science","institution_ids":["https://openalex.org/I100005738"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5064318273","display_name":"Dan Ventura","orcid":"https://orcid.org/0000-0002-3111-2238"},"institutions":[{"id":"https://openalex.org/I100005738","display_name":"Brigham Young University","ror":"https://ror.org/047rhhm47","country_code":"US","type":"education","lineage":["https://openalex.org/I100005738"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Dan Ventura","raw_affiliation_strings":["Brigham Young University,Department of Computer Science"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Brigham Young University,Department of Computer Science","institution_ids":["https://openalex.org/I100005738"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5074307376"],"corresponding_institution_ids":["https://openalex.org/I100005738"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.85983587,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"899","last_page":"904"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.3154999911785126,"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":0.3154999911785126,"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/T12031","display_name":"Speech and dialogue systems","score":0.1761000007390976,"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":0.14010000228881836,"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/language-model","display_name":"Language model","score":0.448199987411499},{"id":"https://openalex.org/keywords/controller","display_name":"Controller (irrigation)","score":0.38940000534057617},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.36649999022483826},{"id":"https://openalex.org/keywords/text-generation","display_name":"Text generation","score":0.3314000070095062},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.30079999566078186},{"id":"https://openalex.org/keywords/language-identification","display_name":"Language identification","score":0.2955000102519989},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.27570000290870667}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6852999925613403},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4853000044822693},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.45969998836517334},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.448199987411499},{"id":"https://openalex.org/C203479927","wikidata":"https://www.wikidata.org/wiki/Q5165939","display_name":"Controller (irrigation)","level":2,"score":0.38940000534057617},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.36649999022483826},{"id":"https://openalex.org/C2985684807","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Text generation","level":2,"score":0.3314000070095062},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.30079999566078186},{"id":"https://openalex.org/C129792486","wikidata":"https://www.wikidata.org/wiki/Q1050419","display_name":"Language identification","level":3,"score":0.2955000102519989},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.27730000019073486},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.27570000290870667},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2628999948501587},{"id":"https://openalex.org/C2776187449","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Natural language generation","level":3,"score":0.26089999079704285},{"id":"https://openalex.org/C2778751112","wikidata":"https://www.wikidata.org/wiki/Q835016","display_name":"Window (computing)","level":2,"score":0.259799987077713},{"id":"https://openalex.org/C115901376","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automation","level":2,"score":0.2574999928474426},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.25189998745918274},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.25110000371932983}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icmla66185.2025.00134","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icmla66185.2025.00134","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Conference on Machine Learning and Applications (ICMLA)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":21,"referenced_works":["https://openalex.org/W2251939518","https://openalex.org/W2736601468","https://openalex.org/W2963548348","https://openalex.org/W2970986510","https://openalex.org/W2971307358","https://openalex.org/W2979826702","https://openalex.org/W3085190015","https://openalex.org/W3098267758","https://openalex.org/W3100355250","https://openalex.org/W3151929433","https://openalex.org/W3176618728","https://openalex.org/W4221144473","https://openalex.org/W4224115290","https://openalex.org/W4231136072","https://openalex.org/W4360764510","https://openalex.org/W4385573003","https://openalex.org/W4385573040","https://openalex.org/W4386290290","https://openalex.org/W4399789779","https://openalex.org/W4410080077","https://openalex.org/W7133211372"],"related_works":[],"abstract_inverted_index":{"As":[0],"the":[1,36,77,100,122,151],"primary":[2],"means":[3],"of":[4,38,102,135,139],"interaction":[5],"with":[6,158],"pretrained":[7,85],"language":[8,46,62,86,105,153],"models":[9],"shifts":[10],"from":[11,76],"local":[12],"to":[13,17,64,93,98,127],"remote":[14],"connection,":[15],"access":[16],"fundamental":[18],"model":[19,63,87,154],"features":[20,75],"such":[21],"as":[22,155],"token":[23],"embeddings,":[24],"hidden":[25],"states,":[26],"and":[27,114,137,162],"output":[28,164],"probabilities":[29],"have":[30],"become":[31],"restricted.":[32],"These":[33],"restrictions":[34],"reduce":[35],"viability":[37],"established":[39],"controllable":[40,129],"text":[41,66,130,141,160,163],"generation":[42,101,131],"methods":[43,53],"for":[44,54],"large":[45],"models.":[47],"To":[48],"address":[49],"this,":[50],"we":[51],"propose":[52],"a":[55,60,68,84,95,103,156],"black-box":[56,81,123],"controller":[57,82,108,124],"that":[58,121],"steers":[59],"base":[61,78,104,152],"generate":[65,94],"possessing":[67],"target":[69],"attribute":[70],"without":[71],"relying":[72],"on":[73,111],"any":[74],"model.":[79,106],"The":[80,107,118],"is":[83,109,125,147],"fine-tuned":[88],"using":[89],"Proximal":[90],"Policy":[91],"Optimization":[92],"control":[96,113],"prefix":[97],"guide":[99],"evaluated":[110],"sentiment":[112],"toxicity":[115],"avoidance":[116],"tasks.":[117],"results":[119],"show":[120],"comparable":[126],"other":[128],"baselines":[132],"in":[133],"terms":[134],"accuracy":[136],"diversity":[138],"generated":[140],"while":[142],"maintaining":[143],"high":[144],"fluency.":[145],"This":[146],"achieved":[148],"despite":[149],"treating":[150],"black-box,":[157],"only":[159],"input":[161],"interaction.":[165]},"counts_by_year":[],"updated_date":"2026-05-05T08:41:31.759640","created_date":"2026-04-08T00:00:00"}