{"id":"https://openalex.org/W7139098629","doi":"https://doi.org/10.48550/arxiv.2603.16459","title":"DynHD: Hallucination Detection for Diffusion Large Language Models via Denoising Dynamics Deviation Learning","display_name":"DynHD: Hallucination Detection for Diffusion Large Language Models via Denoising Dynamics Deviation Learning","publication_year":2026,"publication_date":"2026-03-17","ids":{"openalex":"https://openalex.org/W7139098629","doi":"https://doi.org/10.48550/arxiv.2603.16459"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.16459","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.16459","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.2603.16459","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129902087","display_name":"Yanyu Qian","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Qian, Yanyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129870538","display_name":"Yue Tan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tan, Yue","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129922935","display_name":"Yixin Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yixin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129897704","display_name":"Wang Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Wang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129881361","display_name":"Shirui Pan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pan, Shirui","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5129902087"],"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/T12488","display_name":"Mental Health via Writing","score":0.12680000066757202,"subfield":{"id":"https://openalex.org/subfields/3207","display_name":"Social Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12488","display_name":"Mental Health via Writing","score":0.12680000066757202,"subfield":{"id":"https://openalex.org/subfields/3207","display_name":"Social Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.0868000015616417,"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/T11147","display_name":"Misinformation and Its Impacts","score":0.08470000326633453,"subfield":{"id":"https://openalex.org/subfields/3312","display_name":"Sociology and Political Science"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.5315999984741211},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5277000069618225},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.490200012922287},{"id":"https://openalex.org/keywords/generator","display_name":"Generator (circuit theory)","score":0.4595000147819519},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.41530001163482666},{"id":"https://openalex.org/keywords/detector","display_name":"Detector","score":0.39259999990463257},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3799000084400177}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7461000084877014},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6068999767303467},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.5315999984741211},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5277000069618225},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.490200012922287},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.4595000147819519},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.41530001163482666},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3995000123977661},{"id":"https://openalex.org/C94915269","wikidata":"https://www.wikidata.org/wiki/Q1834857","display_name":"Detector","level":2,"score":0.39259999990463257},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3799000084400177},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.35370001196861267},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.3531999886035919},{"id":"https://openalex.org/C145912823","wikidata":"https://www.wikidata.org/wiki/Q113558","display_name":"Dynamics (music)","level":2,"score":0.3357999920845032},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.32600000500679016},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.3140000104904175},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.29010000824928284},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.27959999442100525},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.27900001406669617}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.16459","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.16459","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.2603.16459","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.16459","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Diffusion":[0],"large":[1],"language":[2],"models":[3,13],"(D-LLMs)":[4],"have":[5],"emerged":[6],"as":[7,45],"a":[8,23,71,133,161,176],"promising":[9],"alternative":[10],"to":[11,15,49,66],"auto-regressive":[12],"due":[14],"their":[16,28],"iterative":[17],"refinement":[18],"capabilities.":[19],"However,":[20],"hallucinations":[21],"remain":[22],"critical":[24],"issue":[25],"that":[26,62,109,138,165,180,196],"hinders":[27],"reliability.":[29],"To":[30,123,152],"detect":[31],"hallucination":[32,67,101,157,178],"responses":[33],"from":[34,113],"model":[35,153],"outputs,":[36],"token-level":[37],"uncertainty":[38,82,172],"(e.g.,":[39],"entropy)":[40],"has":[41],"been":[42],"widely":[43],"used":[44],"an":[46],"effective":[47],"signal":[48],"indicate":[50],"potential":[51],"factual":[52],"errors.":[53],"Nevertheless,":[54],"the":[55,78,84,93,125,167,185,188],"fixed-length":[56],"generation":[57],"paradigm":[58],"of":[59,81,95,171],"D-LLMs":[60],"implies":[61],"tokens":[63,146],"contribute":[64],"unevenly":[65],"detection,":[68,158],"with":[69,175],"only":[70],"small":[72],"subset":[73],"providing":[74],"meaningful":[75,150],"signals.":[76],"Moreover,":[77],"evolution":[79,169],"trend":[80],"throughout":[83],"diffusion":[85],"process":[86],"can":[87],"also":[88],"provide":[89],"important":[90],"signals,":[91],"highlighting":[92],"necessity":[94],"modeling":[96],"its":[97],"denoising":[98,154],"dynamics":[99,155],"for":[100,156],"detection.":[102],"In":[103],"this":[104],"paper,":[105],"we":[106,131,159],"propose":[107,132],"DynHD":[108,197],"bridge":[110],"these":[111],"gaps":[112],"both":[114],"spatial":[115],"(token":[116],"sequence)":[117],"and":[118,147,190,209],"temporal":[119],"(denoising":[120],"dynamics)":[121],"perspectives.":[122],"address":[124],"information":[126],"density":[127],"imbalance":[128],"across":[129,206],"tokens,":[130],"semantic-aware":[134],"evidence":[135,163],"construction":[136],"module":[137],"extracts":[139],"hallucination-indicative":[140],"signals":[141],"by":[142,183],"filtering":[143],"out":[144],"non-informative":[145],"emphasizing":[148],"semantically":[149],"ones.":[151],"introduce":[160],"reference":[162,191],"generator":[164],"learns":[166],"expected":[168],"trajectory":[170],"evidence,":[173],"along":[174],"deviation-based":[177],"detector":[179],"makes":[181],"predictions":[182],"measuring":[184],"discrepancy":[186],"between":[187],"observed":[189],"trajectories.":[192],"Extensive":[193],"experiments":[194],"demonstrate":[195],"consistently":[198],"outperforms":[199],"state-of-the-art":[200],"baselines":[201],"while":[202],"achieving":[203],"higher":[204],"efficiency":[205],"multiple":[207],"benchmarks":[208],"backbone":[210],"models.":[211]},"counts_by_year":[],"updated_date":"2026-03-20T20:54:20.808490","created_date":"2026-03-20T00:00:00"}