{"id":"https://openalex.org/W7202381085","doi":"https://doi.org/10.48550/arxiv.2608.11249","title":"Diffuse to Compress: Leveraging Diffusion LMs for Lossless Compression","display_name":"Diffuse to Compress: Leveraging Diffusion LMs for Lossless Compression","publication_year":2026,"publication_date":"2026-08-04","ids":{"openalex":"https://openalex.org/W7202381085","doi":"https://doi.org/10.48550/arxiv.2608.11249"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.11249","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.11249","pdf_url":"https://arxiv.org/pdf/2608.11249","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","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.11249","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5021065131","display_name":"Angelo Nardone","orcid":"https://orcid.org/0009-0006-2068-5934","observed_orcids":["https://orcid.org/0009-0006-2068-5934"]},"institutions":[{"id":"https://openalex.org/I108290504","display_name":"University of Pisa","ror":"https://ror.org/03ad39j10","country_code":"IT","type":"education","lineage":["https://openalex.org/I108290504"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Nardone, Angelo","raw_affiliation_strings":["University of Pisa , Department of Computer Science , Pisa , Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Pisa , Department of Computer Science , Pisa , Italy","institution_ids":["https://openalex.org/I108290504"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5135435445","display_name":"Paolo Ferragina","orcid":"https://orcid.org/0000-0003-1353-360X","observed_orcids":["https://orcid.org/0000-0003-1353-360X"]},"institutions":[{"id":"https://openalex.org/I108290504","display_name":"University of Pisa","ror":"https://ror.org/03ad39j10","country_code":"IT","type":"education","lineage":["https://openalex.org/I108290504"]},{"id":"https://openalex.org/I162290304","display_name":"Scuola Superiore Sant'Anna","ror":"https://ror.org/025602r80","country_code":"IT","type":"education","lineage":["https://openalex.org/I162290304"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Ferragina, Paolo","raw_affiliation_strings":["Scuola Superiore Sant'Anna , Pisa , Italy","University of Pisa , Department of Computer Science , Pisa , Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Scuola Superiore Sant'Anna , Pisa , Italy","institution_ids":["https://openalex.org/I162290304"]},{"raw_affiliation_string":"University of Pisa , Department of Computer Science , Pisa , Italy","institution_ids":["https://openalex.org/I108290504"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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/T11269","display_name":"Algorithms and Data Compression","score":0.9990000128746033,"subfield":{"id":"https://openalex.org/subfields/2614","display_name":"Theoretical Computer Science"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11269","display_name":"Algorithms and Data Compression","score":0.9990000128746033,"subfield":{"id":"https://openalex.org/subfields/2614","display_name":"Theoretical Computer Science"},"field":{"id":"https://openalex.org/fields/26","display_name":"Mathematics"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10901","display_name":"Advanced Data Compression Techniques","score":0.000699999975040555,"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/T10028","display_name":"Topic Modeling","score":9.999999747378752e-05,"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/lossless-text-compression","display_name":"lossless text compression","score":0.9779999852180481},{"id":"https://openalex.org/keywords/diffusion-language-models","display_name":"diffusion language models","score":0.968999981880188},{"id":"https://openalex.org/keywords/throughput-bottleneck","display_name":"throughput bottleneck","score":0.6600000262260437}],"concepts":[{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.9018999934196472},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7597000002861023},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.7110000252723694},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.6915000081062317},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.5439000129699707},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5152000188827515},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.46779999136924744},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.45980000495910645},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.438400000333786},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4311999976634979},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4000999927520752},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.36469998955726624},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3528999984264374},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3458999991416931},{"id":"https://openalex.org/C165021410","wikidata":"https://www.wikidata.org/wiki/Q55564","display_name":"Lossy compression","level":2,"score":0.34369999170303345},{"id":"https://openalex.org/C94835093","wikidata":"https://www.wikidata.org/wiki/Q3113333","display_name":"Data compression ratio","level":5,"score":0.33070001006126404},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3237999975681305},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.30160000920295715},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.2930999994277954},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.2928999960422516},{"id":"https://openalex.org/C13481523","wikidata":"https://www.wikidata.org/wiki/Q412438","display_name":"Image compression","level":4,"score":0.29089999198913574},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.28790000081062317},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.28279998898506165},{"id":"https://openalex.org/C57890076","wikidata":"https://www.wikidata.org/wiki/Q4680725","display_name":"Adaptive coding","level":4,"score":0.2655999958515167},{"id":"https://openalex.org/C25797200","wikidata":"https://www.wikidata.org/wiki/Q828137","display_name":"Compression ratio","level":3,"score":0.2648000121116638},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2632000148296356},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.25929999351501465},{"id":"https://openalex.org/C160145156","wikidata":"https://www.wikidata.org/wiki/Q778586","display_name":"Executable","level":2,"score":0.25839999318122864}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.11249","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.11249","pdf_url":"https://arxiv.org/pdf/2608.11249","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.2608.11249","doi":"https://doi.org/10.48550/arxiv.2608.11249","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.11249","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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.11249","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.11249","pdf_url":"https://arxiv.org/pdf/2608.11249","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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7202381085.pdf","grobid_xml":"https://content.openalex.org/works/W7202381085.grobid-xml"},"referenced_works_count":9,"referenced_works":["https://openalex.org/W1933799648","https://openalex.org/W2901125983","https://openalex.org/W4379933437","https://openalex.org/W4401306886","https://openalex.org/W4403780929","https://openalex.org/W4403785118","https://openalex.org/W4403964524","https://openalex.org/W4417297081","https://openalex.org/W7171546479"],"related_works":["https://openalex.org/W4403785118","https://openalex.org/W4403780929","https://openalex.org/W7214959833","https://openalex.org/W4408854911","https://openalex.org/W4394007739","https://openalex.org/W4417511804","https://openalex.org/W4410007609","https://openalex.org/W6922767553","https://openalex.org/W7047617252","https://openalex.org/W4412934273"],"abstract_inverted_index":{"We":[0,121,178],"study":[1],"the":[2,10,14,95,99,130,136,160,163,207,213,216],"problem":[3],"of":[4,18,101,167,215],"lossless":[5,102,157,219],"text":[6,76,104,220],"compression,":[7,105,158],"motivated":[8],"by":[9,36,140,153],"rapid":[11],"growth":[12],"in":[13,39,98,218],"collection":[15],"and":[16,28,35,77,165,181,188,194,236],"storage":[17],"digital":[19],"textual":[20,201],"data":[21],"-":[22,34],"including":[23],"plain":[24],"text,":[25],"source":[26],"code,":[27],"structured":[29],"formats":[30],"such":[31,69],"as":[32,70,112,223],"XML":[33],"recent":[37,46,231],"advances":[38,212,232],"neural":[40,81,103],"language":[41],"model-based":[42],"compression.":[43,221],"In":[44],"particular,":[45],"LLM-based":[47,119,193],"approaches,":[48],"whether":[49],"built":[50],"on":[51,75,197],"symbol-ranking":[52],"pipelines":[53],"or":[54,73],"paired":[55],"with":[56,127],"a":[57,199,227],"statistical":[58],"compressor,":[59],"have":[60],"demonstrated":[61],"compression":[62,132],"ratios":[63],"significantly":[64],"superior":[65],"to":[66,117,156,174,184],"general-purpose":[67,195],"compressors":[68,196],"zstd,":[71],"gzip,":[72],"bzip":[74],"code.":[78],"However,":[79,144],"these":[80,146,186],"approaches":[82],"suffer":[83],"from":[84],"severe":[85],"throughput":[86,137],"limitations,":[87],"making":[88],"them":[89,190],"not":[90],"yet":[91],"practically":[92],"usable.":[93],"For":[94],"first":[96],"time":[97],"context":[100],"we":[106],"introduce":[107],"Diffusion":[108],"Language":[109],"Models":[110],"(DLMs)":[111],"an":[113],"alternative":[114],"inference":[115],"paradigm":[116],"autoregressive":[118,125],"approaches.":[120],"argue":[122],"that":[123,206],"replacing":[124],"LLMs":[126],"DLMs":[128,155,224],"within":[129],"same":[131],"framework":[133,211],"could":[134],"overcome":[135],"bottleneck":[138],"caused":[139],"their":[141],"one-symbol-per-step":[142],"limitation.":[143],"achieving":[145],"improvements":[147],"requires":[148],"addressing":[149],"algorithmic":[150],"challenges":[151,187],"introduced":[152],"applying":[154],"where":[159],"architecture":[161],"allows":[162],"number":[164],"positions":[166],"symbols":[168],"encoded":[169],"at":[170],"each":[171],"forward":[172],"pass":[173],"be":[175],"decided":[176],"independently.":[177],"design":[179],"efficient":[180,237],"effective":[182],"strategies":[183],"solve":[185],"evaluate":[189],"experimentally":[191],"against":[192],"enwik8,":[198],"well-established":[200],"benchmark.":[202],"Our":[203],"results":[204],"show":[205],"newly":[208],"proposed":[209],"DLM-based":[210],"state":[214],"art":[217],"Moreover,":[222],"are":[225],"still":[226],"relatively":[228],"young":[229],"paradigm,":[230],"toward":[233],"increasingly":[234],"capable":[235],"models":[238],"suggest":[239],"substantial":[240],"room":[241],"for":[242],"further":[243],"improvements.":[244]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-08-14T00:00:00"}