{"id":"https://openalex.org/W7163183471","doi":"https://doi.org/10.48550/arxiv.2606.01926","title":"Mitigating Bias in Locally Constrained Decoding via Tractable Proposals","display_name":"Mitigating Bias in Locally Constrained Decoding via Tractable Proposals","publication_year":2026,"publication_date":"2026-06-01","ids":{"openalex":"https://openalex.org/W7163183471","doi":"https://doi.org/10.48550/arxiv.2606.01926"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.01926","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01926","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.01926","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5043265692","display_name":"Meihua Dang","orcid":"https://orcid.org/0000-0001-8241-1943"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dang, Meihua","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137622840","display_name":"Linxin Song","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Linxin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018876172","display_name":"Honghua Zhang","orcid":"https://orcid.org/0000-0003-1228-5616"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Honghua","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137712295","display_name":"Jieyu Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Jieyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137672622","display_name":"Guy Van den Broeck","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Broeck, Guy Van den","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137708322","display_name":"Stefano Ermon","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ermon, Stefano","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":0.4361000061035156,"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":0.4361000061035156,"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":0.13740000128746033,"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":0.1143999993801117,"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/decoding-methods","display_name":"Decoding methods","score":0.6974999904632568},{"id":"https://openalex.org/keywords/probabilistic-logic","display_name":"Probabilistic logic","score":0.6319000124931335},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.4796000123023987},{"id":"https://openalex.org/keywords/importance-sampling","display_name":"Importance sampling","score":0.4496000111103058},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.40860000252723694},{"id":"https://openalex.org/keywords/probability-distribution","display_name":"Probability distribution","score":0.39469999074935913},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.3709999918937683},{"id":"https://openalex.org/keywords/automaton","display_name":"Automaton","score":0.37040001153945923},{"id":"https://openalex.org/keywords/markov-chain-monte-carlo","display_name":"Markov chain Monte Carlo","score":0.35440000891685486},{"id":"https://openalex.org/keywords/finite-state-machine","display_name":"Finite-state machine","score":0.35370001196861267}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7265999913215637},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.6974999904632568},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.6319000124931335},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.5454000234603882},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.4796000123023987},{"id":"https://openalex.org/C52740198","wikidata":"https://www.wikidata.org/wiki/Q1539564","display_name":"Importance sampling","level":3,"score":0.4496000111103058},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4327000081539154},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.40860000252723694},{"id":"https://openalex.org/C149441793","wikidata":"https://www.wikidata.org/wiki/Q200726","display_name":"Probability distribution","level":2,"score":0.39469999074935913},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.3709999918937683},{"id":"https://openalex.org/C112505250","wikidata":"https://www.wikidata.org/wiki/Q787116","display_name":"Automaton","level":2,"score":0.37040001153945923},{"id":"https://openalex.org/C111350023","wikidata":"https://www.wikidata.org/wiki/Q1191869","display_name":"Markov chain Monte Carlo","level":3,"score":0.35440000891685486},{"id":"https://openalex.org/C167822520","wikidata":"https://www.wikidata.org/wiki/Q176452","display_name":"Finite-state machine","level":2,"score":0.35370001196861267},{"id":"https://openalex.org/C159886148","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov process","level":2,"score":0.33480000495910645},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3343999981880188},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.32739999890327454},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.32499998807907104},{"id":"https://openalex.org/C98763669","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov chain","level":2,"score":0.3158000111579895},{"id":"https://openalex.org/C158424031","wikidata":"https://www.wikidata.org/wiki/Q1191905","display_name":"Gibbs sampling","level":3,"score":0.29820001125335693},{"id":"https://openalex.org/C2777402240","wikidata":"https://www.wikidata.org/wiki/Q6783436","display_name":"Masking (illustration)","level":2,"score":0.2939999997615814},{"id":"https://openalex.org/C204397858","wikidata":"https://www.wikidata.org/wiki/Q4437907","display_name":"List decoding","level":5,"score":0.29350000619888306},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2930999994277954},{"id":"https://openalex.org/C149189445","wikidata":"https://www.wikidata.org/wiki/Q5283894","display_name":"Divergence-from-randomness model","level":3,"score":0.29030001163482666},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.2890999913215637},{"id":"https://openalex.org/C162500139","wikidata":"https://www.wikidata.org/wiki/Q2835887","display_name":"Estimation of distribution algorithm","level":2,"score":0.28839999437332153},{"id":"https://openalex.org/C193969084","wikidata":"https://www.wikidata.org/wiki/Q7452500","display_name":"Sequential decoding","level":4,"score":0.2870999872684479},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.28630000352859497},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.2741999924182892},{"id":"https://openalex.org/C19499675","wikidata":"https://www.wikidata.org/wiki/Q232207","display_name":"Monte Carlo method","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.26669999957084656},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.26499998569488525},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.26339998841285706},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.2615000009536743},{"id":"https://openalex.org/C2780416260","wikidata":"https://www.wikidata.org/wiki/Q2063","display_name":"JSON","level":2,"score":0.2515999972820282}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.01926","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01926","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.01926","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.01926","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":"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":{"Generations":[0],"from":[1,78],"large":[2],"language":[3],"models":[4],"often":[5],"fail":[6],"to":[7,9,46,70,103,131,172,178],"conform":[8],"desired":[10],"constraints":[11,23,87],"such":[12,48],"as":[13,89,124],"JSON":[14],"schema.":[15],"Existing":[16],"locally":[17],"constrained":[18,106],"decoding":[19,107],"(LCD)":[20],"approaches":[21],"enforce":[22],"by":[24],"myopically":[25],"masking":[26],"out":[27],"next":[28],"tokens,":[29],"resulting":[30],"in":[31,36],"biased":[32],"sampling":[33,77,169],"and":[34,73,141,159],"degradation":[35],"performance.":[37],"Recent":[38],"work":[39],"uses":[40],"sequential":[41],"Monte":[42],"Carlo":[43],"(SMC)":[44],"methods":[45],"mitigate":[47],"biases,":[49],"but":[50],"designing":[51],"effective":[52],"proposal":[53],"distributions":[54],"or":[55],"potential":[56],"functions":[57],"remains":[58],"a":[59,67],"key":[60],"challenge.":[61],"In":[62,110],"this":[63],"work,":[64],"we":[65,84,101,128],"propose":[66],"generic":[68],"approach":[69],"construct":[71,104],"proposals":[72,137],"potentials":[74],"for":[75,95],"SMC":[76,168],"$p_{\\mathrm{lm}}(":[79],"\\cdot":[80],"\\mid":[81],"\\mathrm{constraint})$.":[82],"First,":[83],"show":[85,163],"that":[86,115,164],"specified":[88],"finite":[90,117],"automata":[91,118],"can":[92],"be":[93],"tensorized":[94,116],"efficient":[96],"execution":[97],"on":[98,151],"GPUs,":[99],"which":[100],"use":[102],"globally":[105],"(GCD)":[108],"proposals.":[109],"addition,":[111],"leveraging":[112],"the":[113,120,133,145,152,166,179],"fact":[114],"share":[119],"same":[121,167],"circuit":[122],"structure":[123],"hidden":[125],"Markov":[126],"models,":[127],"circuit-multiply":[129],"them":[130],"obtain":[132],"probabilistic":[134,142],"GCD":[135],"(P-GCD)":[136],"encoding":[138],"both":[139],"logical":[140],"information":[143],"about":[144],"target":[146,180],"distributions.":[147],"We":[148],"evaluate":[149],"(P-)GCD":[150,175],"tasks":[153],"of":[154],"function":[155],"calling,":[156],"keyword-based":[157],"generation,":[158],"SQL":[160],"generation.":[161],"Experiments":[162],"under":[165],"setup,":[170],"compared":[171],"LCD":[173],"proposals,":[174],"converges":[176],"faster":[177],"distribution":[181],"with":[182],"significantly":[183],"fewer":[184],"particles.":[185]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-03T00:00:00"}