{"id":"https://openalex.org/W7161172251","doi":"https://doi.org/10.48550/arxiv.2605.12825","title":"Orthrus: Memory-Efficient Parallel Token Generation via Dual-View Diffusion","display_name":"Orthrus: Memory-Efficient Parallel Token Generation via Dual-View Diffusion","publication_year":2026,"publication_date":"2026-05-12","ids":{"openalex":"https://openalex.org/W7161172251","doi":"https://doi.org/10.48550/arxiv.2605.12825"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.12825","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12825","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,"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.2605.12825","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128192991","display_name":"Chien Nguyen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Van Nguyen, Chien","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084795440","display_name":"Chaitra Hegde","orcid":"https://orcid.org/0000-0002-2791-4254"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hegde, Chaitra","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136132857","display_name":"Van Cuong Pham","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pham, Van Cuong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136153378","display_name":"Ryan A. Rossi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rossi, Ryan A.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028863551","display_name":"Franck Dernoncourt","orcid":"https://orcid.org/0000-0002-1119-1346"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dernoncourt, Franck","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136180075","display_name":"Thien Huu Nguyen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nguyen, Thien Huu","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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.439300000667572,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.439300000667572,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.09380000084638596,"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":0.07660000026226044,"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/autoregressive-model","display_name":"Autoregressive model","score":0.6901000142097473},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5659999847412109},{"id":"https://openalex.org/keywords/lossless-compression","display_name":"Lossless compression","score":0.5541999936103821},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.5242999792098999},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.505299985408783},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.4814999997615814},{"id":"https://openalex.org/keywords/fidelity","display_name":"Fidelity","score":0.4690999984741211},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.46000000834465027},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.4447999894618988}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7864000201225281},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.6901000142097473},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5659999847412109},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.5541999936103821},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5331000089645386},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.5242999792098999},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.505299985408783},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.4814999997615814},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.4690999984741211},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4627000093460083},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.46000000834465027},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4447999894618988},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.44269999861717224},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.4065000116825104},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3752000033855438},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.3743000030517578},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.3447999954223633},{"id":"https://openalex.org/C143724316","wikidata":"https://www.wikidata.org/wiki/Q312468","display_name":"Series (stratigraphy)","level":2,"score":0.3165000081062317},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.31299999356269836},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.3034000098705292},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.29919999837875366},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.2971000075340271},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.2948000133037567},{"id":"https://openalex.org/C198082294","wikidata":"https://www.wikidata.org/wiki/Q3399648","display_name":"Position (finance)","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C113364801","wikidata":"https://www.wikidata.org/wiki/Q26674","display_name":"High fidelity","level":2,"score":0.27649998664855957},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.26100000739097595}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.12825","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12825","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,"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.2605.12825","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12825","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,"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":{"We":[0],"introduce":[1],"Orthrus,":[2],"a":[3,38,66,87,91,97,156],"simple":[4],"and":[5,65,166],"efficient":[6],"dual-architecture":[7],"framework":[8,85],"that":[9],"unifies":[10],"the":[11,22,84,102,114,121,133,146],"exact":[12,115,142],"generation":[13,26],"fidelity":[14],"of":[15,27,33,68],"autoregressive":[16,35,104,122],"Large":[17],"Language":[18],"Models":[19],"(LLMs)":[20],"with":[21,90,159],"high-speed":[23],"parallel":[24,54,98,137],"token":[25],"diffusion":[28,45,99,134],"models.":[29],"The":[30],"sequential":[31],"nature":[32],"standard":[34,103],"decoding":[36],"represents":[37],"fundamental":[39],"bottleneck":[40],"for":[41],"high-throughput":[42],"inference.":[43],"While":[44],"language":[46],"models":[47],"attempt":[48],"to":[49,78,95,113,127,155],"break":[50],"this":[51,74,107],"barrier":[52],"via":[53],"generation,":[55],"they":[56],"suffer":[57],"from":[58],"significant":[59],"performance":[60],"degradation,":[61],"high":[62],"training":[63],"costs,":[64],"lack":[67],"rigorous":[69],"convergence":[70],"guarantees.":[71],"Orthrus":[72,149],"resolves":[73],"dichotomy":[75],"natively.":[76],"Designed":[77],"seamlessly":[79],"integrate":[80],"into":[81],"existing":[82],"Transformers,":[83],"augments":[86],"frozen":[88],"LLM":[89],"lightweight,":[92],"trainable":[93],"module":[94],"create":[96],"view":[100],"alongside":[101],"view.":[105],"In":[106],"unified":[108],"system,":[109],"both":[110],"views":[111],"attend":[112],"same":[116],"high-fidelity":[117],"Key-Value":[118],"(KV)":[119],"cache;":[120],"head":[123,135],"executes":[124,136],"context":[125],"pre-filling":[126],"construct":[128],"accurate":[129],"KV":[130],"representations,":[131],"while":[132],"generation.":[138],"By":[139],"employing":[140],"an":[141,161],"consensus":[143],"mechanism":[144],"between":[145],"two":[147],"views,":[148],"guarantees":[150],"lossless":[151],"inference,":[152],"delivering":[153],"up":[154],"7.8x":[157],"speedup":[158],"only":[160],"O(1)":[162],"memory":[163],"cache":[164],"overhead":[165],"minimal":[167],"parameter":[168],"additions.":[169]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-05-15T00:00:00"}