{"id":"https://openalex.org/W7155191333","doi":"https://doi.org/10.48550/arxiv.2604.19087","title":"OLLM: Options-based Large Language Models","display_name":"OLLM: Options-based Large Language Models","publication_year":2026,"publication_date":"2026-04-21","ids":{"openalex":"https://openalex.org/W7155191333","doi":"https://doi.org/10.48550/arxiv.2604.19087"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.19087","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.19087","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.19087","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5103511695","display_name":"Shashank Sharma","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Sharma, Shashank","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128366319","display_name":"Janina Hoffmann","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hoffmann, Janina","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5134285715","display_name":"Vinay Namboodiri","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Namboodiri, Vinay","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5103511695"],"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.2071000039577484,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},"topics":[{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.2071000039577484,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.1386999934911728,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.12250000238418579,"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":0.7250999808311462},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.6230000257492065},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5791000127792358},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5386999845504761},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.5284000039100647},{"id":"https://openalex.org/keywords/latent-variable","display_name":"Latent variable","score":0.48590001463890076},{"id":"https://openalex.org/keywords/variation","display_name":"Variation (astronomy)","score":0.44600000977516174},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.39419999718666077}],"concepts":[{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.7250999808311462},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6662999987602234},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6230000257492065},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5813000202178955},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5791000127792358},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5386999845504761},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.5284000039100647},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.48590001463890076},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4796000123023987},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.44600000977516174},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.39419999718666077},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.3935000002384186},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.28119999170303345},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.28029999136924744},{"id":"https://openalex.org/C65965080","wikidata":"https://www.wikidata.org/wiki/Q1806885","display_name":"Latent variable model","level":3,"score":0.26499998569488525},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.2603999972343445},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.26019999384880066},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.25619998574256897},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2529999911785126},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.19087","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.19087","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.19087","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.19087","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"sustainable_development_goals":[{"score":0.4739104211330414,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0,99,141],"introduce":[1],"Options":[2],"LLM":[3,91],"(OLLM),":[4],"a":[5,19,30,49,65,71,81,103,144,159,226],"simple,":[6],"general":[7],"method":[8],"that":[9,74,151,208],"replaces":[10],"the":[11,25,84,148,181],"single":[12],"next-token":[13,56,210],"prediction":[14],"of":[15,21,35,108],"standard":[16],"LLMs":[17],"with":[18,95],"\\textit{set":[20],"learned":[22,187],"options}":[23],"for":[24,229],"next":[26],"token,":[27],"indexed":[28],"by":[29,64],"discrete":[31],"latent":[32,51,139,149],"variable.":[33],"Instead":[34],"relying":[36],"on":[37,112,116],"temperature":[38],"or":[39,62,177,201],"sampling":[40],"heuristics":[41],"to":[42,92,102,134,154,185],"induce":[43],"diversity,":[44],"OLLM":[45,69,101],"models":[46],"variation":[47],"explicitly:":[48],"small":[50],"space":[52,150,162],"parametrizes":[53],"multiple":[54],"plausible":[55],"options":[57,186],"which":[58],"can":[59],"be":[60,93],"selected":[61],"searched":[63],"downstream":[66],"policy.":[67],"Architecturally,":[68],"is":[70,183],"lightweight":[72],"\"plug-in\"":[73],"inserts":[75],"two":[76],"layers:":[77],"an":[78],"encoder":[79],"and":[80,114,169,215,220],"decoder,":[82],"before":[83],"output":[85],"head,":[86],"allowing":[87],"almost":[88],"any":[89],"pretrained":[90],"converted":[94],"minimal":[96],"additional":[97,199],"parameters.":[98],"apply":[100],"1.7B-parameter":[104],"backbone":[105],"(only":[106],"$1.56\\%$":[107],"parameters":[109],"trainable)":[110],"trained":[111],"OpenMathReasoning":[113],"evaluated":[115],"OmniMath.":[117],"The":[118],"SOTA":[119],"LoRA-adapted":[120],"baselines":[121],"peak":[122],"at":[123],"$51\\%$":[124],"final":[125],"answer":[126],"correctness,":[127],"while":[128],"OLLM's":[129],"option":[130,161],"set":[131],"allows":[132],"up":[133],"$\\sim":[135],"70\\%$":[136],"under":[137],"optimal":[138],"selection.":[140],"then":[142],"train":[143],"compact":[145],"policy":[146,182,223],"in":[147,158,217,232],"emits":[152],"latents":[153],"control":[155],"generation.":[156],"Operating":[157],"low-dimensional":[160],"makes":[163],"reward":[164],"optimization":[165],"far":[166],"more":[167],"sample-efficient":[168],"substantially":[170],"reduces":[171],"common":[172],"misalignments":[173],"(e.g.,":[174],"language":[175],"switching":[176],"degenerate":[178],"reasoning),":[179],"as":[180,225],"constrained":[184],"during":[188],"SFT.":[189],"Crucially,":[190],"this":[191],"alignment":[192,203],"arises":[193],"from":[194],"model":[195],"structure":[196],"rather":[197],"than":[198],"KL":[200],"handcrafted":[202],"losses.":[204],"Our":[205],"results":[206],"demonstrate":[207],"optionized":[209],"modeling":[211],"enhances":[212],"controllability,":[213],"robustness,":[214],"efficiency":[216],"math":[218],"reasoning,":[219],"highlight":[221],"latent-space":[222],"learning":[224,231],"promising":[227],"direction":[228],"reinforcement":[230],"LLMs.":[233]},"counts_by_year":[],"updated_date":"2026-04-23T06:20:18.424754","created_date":"2026-04-23T00:00:00"}