{"id":"https://openalex.org/W7150865249","doi":"https://doi.org/10.48550/arxiv.2604.02560","title":"Dependency-Guided Parallel Decoding in Discrete Diffusion Language Models","display_name":"Dependency-Guided Parallel Decoding in Discrete Diffusion Language Models","publication_year":2026,"publication_date":"2026-04-02","ids":{"openalex":"https://openalex.org/W7150865249","doi":"https://doi.org/10.48550/arxiv.2604.02560"},"language":null,"primary_location":{"id":"pmh:oai:arXiv.org:2604.02560","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2604.02560","pdf_url":"https://arxiv.org/pdf/2604.02560","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,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2604.02560","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5093854356","display_name":"Liran Ringel","orcid":"https://orcid.org/0009-0006-5822-6856","observed_orcids":["https://orcid.org/0009-0006-5822-6856"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ringel, Liran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133001592","display_name":"Ameen Ali","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ali, Ameen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5041723842","display_name":"Yaniv Romano","orcid":"https://orcid.org/0000-0001-8951-8583","observed_orcids":["https://orcid.org/0000-0001-8951-8583","https://orcid.org/0000-0001-9031-8829"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Romano, Yaniv","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":true,"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":0.6929000020027161,"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":0.6929000020027161,"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":0.26460000872612,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.01119999960064888,"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/parallel-decoding","display_name":"parallel decoding","score":0.9760000109672546},{"id":"https://openalex.org/keywords/total-variation-distance","display_name":"total variation distance","score":0.7120000123977661}],"concepts":[{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.6797999739646912},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6478000283241272},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.6039999723434448},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.5945000052452087},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.5932000279426575},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.5357000231742859},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.4943000078201294},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.4796000123023987},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.47769999504089355},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.4584999978542328},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.4154999852180481},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.3865000009536743},{"id":"https://openalex.org/C51823790","wikidata":"https://www.wikidata.org/wiki/Q504353","display_name":"Greedy algorithm","level":2,"score":0.38100001215934753},{"id":"https://openalex.org/C44492722","wikidata":"https://www.wikidata.org/wiki/Q327069","display_name":"Conditional probability","level":2,"score":0.37540000677108765},{"id":"https://openalex.org/C90673727","wikidata":"https://www.wikidata.org/wiki/Q901718","display_name":"Product (mathematics)","level":2,"score":0.3601999878883362},{"id":"https://openalex.org/C146834321","wikidata":"https://www.wikidata.org/wiki/Q2979672","display_name":"Closure (psychology)","level":2,"score":0.3537999987602234},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.335999995470047},{"id":"https://openalex.org/C18653775","wikidata":"https://www.wikidata.org/wiki/Q1333358","display_name":"Joint probability distribution","level":2,"score":0.3271999955177307},{"id":"https://openalex.org/C18555067","wikidata":"https://www.wikidata.org/wiki/Q8375051","display_name":"Joint (building)","level":2,"score":0.32600000500679016},{"id":"https://openalex.org/C155846161","wikidata":"https://www.wikidata.org/wiki/Q1143367","display_name":"Graphical model","level":2,"score":0.32030001282691956},{"id":"https://openalex.org/C114928208","wikidata":"https://www.wikidata.org/wiki/Q17104952","display_name":"Pairwise independence","level":5,"score":0.3190000057220459},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.28279998898506165},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.28060001134872437},{"id":"https://openalex.org/C149441793","wikidata":"https://www.wikidata.org/wiki/Q200726","display_name":"Probability distribution","level":2,"score":0.28029999136924744},{"id":"https://openalex.org/C62354387","wikidata":"https://www.wikidata.org/wiki/Q875399","display_name":"Boundary (topology)","level":2,"score":0.2799000144004822},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.2685000002384186},{"id":"https://openalex.org/C4069607","wikidata":"https://www.wikidata.org/wiki/Q868732","display_name":"Aliasing","level":3,"score":0.26669999957084656},{"id":"https://openalex.org/C75291252","wikidata":"https://www.wikidata.org/wiki/Q1315756","display_name":"TRACE (psycholinguistics)","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.2565999925136566},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.25609999895095825},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.25450000166893005},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.2508000135421753}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2604.02560","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2604.02560","pdf_url":"https://arxiv.org/pdf/2604.02560","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,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2604.02560","doi":"https://doi.org/10.48550/arxiv.2604.02560","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.02560","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,"listed_in":[],"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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2604.02560","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2604.02560","pdf_url":"https://arxiv.org/pdf/2604.02560","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,"listed_in":[],"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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G7025107089","display_name":null,"funder_award_id":"729/21","funder_id":"https://openalex.org/F4320322252","funder_display_name":"Israel Science Foundation"},{"id":"https://openalex.org/G7308213304","display_name":null,"funder_award_id":"101163414","funder_id":"https://openalex.org/F4320320300","funder_display_name":"European Commission"}],"funders":[{"id":"https://openalex.org/F4320320300","display_name":"European Commission","ror":"https://ror.org/00k4n6c32"},{"id":"https://openalex.org/F4320322252","display_name":"Israel Science Foundation","ror":"https://ror.org/04sazxf24"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7150865249.pdf","grobid_xml":"https://content.openalex.org/works/W7150865249.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Discrete":[0],"diffusion":[1],"language":[2],"models":[3],"(dLLMs)":[4],"accelerate":[5],"text":[6],"generation":[7],"by":[8],"unmasking":[9],"multiple":[10],"tokens":[11,40],"in":[12],"parallel.":[13],"However,":[14],"parallel":[15,106],"decoding":[16],"introduces":[17],"a":[18,27,49,61,64,79,93],"distributional":[19],"mismatch:":[20],"it":[21,68],"approximates":[22],"the":[23,56,100,109],"joint":[24],"conditional":[25,71],"using":[26],"fully":[28],"factorized":[29],"product":[30],"of":[31,60],"per-token":[32],"marginals,":[33],"which":[34],"degrades":[35],"output":[36],"quality":[37],"when":[38],"selected":[39],"are":[41],"strongly":[42],"dependent.":[43],"We":[44],"propose":[45],"DEMASK":[46,113],"(DEpendency-guided":[47],"unMASKing),":[48],"lightweight":[50],"dependency":[51,88],"predictor":[52],"that":[53],"attaches":[54],"to":[55,125],"final":[57],"hidden":[58],"states":[59],"dLLM.":[62],"In":[63],"single":[65],"forward":[66],"pass,":[67],"estimates":[69],"pairwise":[70],"influences":[72],"between":[73,104],"masked":[74],"positions.":[75],"Using":[76],"these":[77],"predictions,":[78],"greedy":[80],"selection":[81],"algorithm":[82],"identifies":[83],"positions":[84],"with":[85],"bounded":[86],"cumulative":[87],"for":[89],"simultaneous":[90],"unmasking.":[91],"Under":[92],"sub-additivity":[94],"assumption,":[95],"we":[96],"prove":[97],"this":[98],"bounds":[99],"total":[101],"variation":[102],"distance":[103],"our":[105],"sampling":[107],"and":[108,127],"model's":[110],"joint.":[111],"Empirically,":[112],"achieves":[114],"1.7-2.2$\\times$":[115],"speedup":[116],"on":[117],"Dream-7B":[118],"while":[119],"matching":[120],"or":[121],"improving":[122],"accuracy":[123],"compared":[124],"confidence-based":[126],"KL-based":[128],"baselines.":[129]},"counts_by_year":[],"updated_date":"2026-09-30T09:27:27.532063","created_date":"2026-04-07T00:00:00"}