{"id":"https://openalex.org/W7161116138","doi":"https://doi.org/10.48550/arxiv.2605.12649","title":"DIVER:Diving Deeper into Distilled Data via Expressive Semantic Recovery","display_name":"DIVER:Diving Deeper into Distilled Data via Expressive Semantic Recovery","publication_year":2026,"publication_date":"2026-05-12","ids":{"openalex":"https://openalex.org/W7161116138","doi":"https://doi.org/10.48550/arxiv.2605.12649"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.12649","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12649","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.2605.12649","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5008716657","display_name":"Qianxin Xia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xia, Qianxin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136176823","display_name":"Zhiyong Shu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shu, Zhiyong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136175684","display_name":"Wenbo Jiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Wenbo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136120894","display_name":"Jiawei Du","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Du, Jiawei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043163390","display_name":"Jielei Wang","orcid":"https://orcid.org/0000-0003-2882-7053"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jielei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136179698","display_name":"Guoming Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Guoming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":6,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.48510000109672546,"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/T10764","display_name":"Privacy-Preserving Technologies in Data","score":0.48510000109672546,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.08940000087022781,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.08139999955892563,"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/semantics","display_name":"Semantics (computer science)","score":0.6353999972343445},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.5546000003814697},{"id":"https://openalex.org/keywords/ambiguity","display_name":"Ambiguity","score":0.4512999951839447},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.4325000047683716},{"id":"https://openalex.org/keywords/overfitting","display_name":"Overfitting","score":0.38499999046325684},{"id":"https://openalex.org/keywords/semantic-data-model","display_name":"Semantic data model","score":0.38100001215934753},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.3628999888896942},{"id":"https://openalex.org/keywords/semantic-web","display_name":"Semantic Web","score":0.33329999446868896}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7906000018119812},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.6353999972343445},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5882999897003174},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.5546000003814697},{"id":"https://openalex.org/C2780522230","wikidata":"https://www.wikidata.org/wiki/Q1140419","display_name":"Ambiguity","level":2,"score":0.4512999951839447},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4505000114440918},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4325000047683716},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.38499999046325684},{"id":"https://openalex.org/C90312973","wikidata":"https://www.wikidata.org/wiki/Q7449052","display_name":"Semantic data model","level":2,"score":0.38100001215934753},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3718999922275543},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3628999888896942},{"id":"https://openalex.org/C2129575","wikidata":"https://www.wikidata.org/wiki/Q54837","display_name":"Semantic Web","level":2,"score":0.33329999446868896},{"id":"https://openalex.org/C511149849","wikidata":"https://www.wikidata.org/wiki/Q7449051","display_name":"Semantic computing","level":3,"score":0.32019999623298645},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.31049999594688416},{"id":"https://openalex.org/C197914299","wikidata":"https://www.wikidata.org/wiki/Q18650","display_name":"Semantic memory","level":3,"score":0.29600000381469727},{"id":"https://openalex.org/C2779803651","wikidata":"https://www.wikidata.org/wiki/Q5282088","display_name":"Discriminator","level":3,"score":0.2840999960899353},{"id":"https://openalex.org/C118930307","wikidata":"https://www.wikidata.org/wiki/Q600590","display_name":"Tuple","level":2,"score":0.2736999988555908},{"id":"https://openalex.org/C2778493491","wikidata":"https://www.wikidata.org/wiki/Q7449072","display_name":"Semantic matching","level":3,"score":0.27079999446868896},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.26649999618530273},{"id":"https://openalex.org/C85407183","wikidata":"https://www.wikidata.org/wiki/Q1045785","display_name":"Semantic network","level":2,"score":0.262800008058548},{"id":"https://openalex.org/C67277372","wikidata":"https://www.wikidata.org/wiki/Q7449085","display_name":"Semantic role labeling","level":3,"score":0.25699999928474426},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2524000108242035}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.12649","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12649","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.2605.12649","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.12649","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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Dataset":[0],"distillation":[1,32,69,176],"aims":[2],"to":[3,54,79,111,141,187],"synthesize":[4],"a":[5,30,43,66],"compact":[6],"proxy":[7],"dataset":[8,17],"that":[9,40],"is":[10,139],"unreadable":[11],"or":[12],"non-raw":[13],"from":[14,36],"the":[15,48,75,108,118,125,128,133,147,151,161,167],"original":[16,129],"for":[18],"privacy":[19],"protection":[20],"and":[21,52,96,116,157,169,178],"highly":[22],"efficient":[23],"learning.":[24],"However,":[25],"previous":[26],"approaches":[27],"typically":[28],"adopt":[29],"single-stage":[31],"paradigm,":[33],"which":[34,73],"suffers":[35],"learning":[37],"specific":[38],"patterns":[39],"overfit":[41],"on":[42,190],"prior":[44],"architecture,":[45],"consequently":[46],"suppressing":[47],"expression":[49],"of":[50,92,103,127,150,171,197],"semantics":[51,102,130],"leading":[53],"performance":[55],"degradation":[56],"across":[57],"heterogeneous":[58],"architectures.":[59],"To":[60],"address":[61],"this":[62],"issue,":[63],"we":[64],"propose":[65],"novel":[67],"dual-stage":[68],"framework":[70],"called":[71],"${\\textbf{DIVER}}$,":[72],"leverages":[74],"pre-trained":[76],"diffusion":[77],"model":[78],"dive":[80],"deeper":[81],"into":[82,107],"$\\textbf{DI}$stilled":[83],"data":[84],"$\\textbf{V}$ia":[85],"$\\textbf{E}$xpressive":[86],"semantic":[87,93,122,137,143,155],"$\\textbf{R}$ecovery,":[88],"an":[89],"entire":[90],"process":[91],"inheritance,":[94],"guidance,":[95],"fusion.":[97],"Semantic":[98],"inheritance":[99],"distills":[100],"high-level":[101],"abstract":[104],"distilled":[105],"images":[106],"latent":[109],"space":[110],"filter":[112],"out":[113],"architecture-specific":[114],"``noise\"":[115],"retain":[117],"intrinsic":[119],"semantics.":[120],"Furthermore,":[121],"guidance":[123,144,162],"improves":[124],"preservation":[126],"by":[131],"directing":[132],"reverse":[134,152],"procedure.":[135],"Finally,":[136],"fusion":[138],"designed":[140],"provide":[142],"only":[145,194],"during":[146],"concrete":[148],"phase":[149],"process,":[153],"preventing":[154],"ambiguity":[156],"artifacts":[158],"while":[159],"maintaining":[160],"information.":[163],"Extensive":[164],"experiments":[165],"validate":[166],"effectiveness":[168],"efficiency":[170],"DIVER":[172],"in":[173],"improving":[174,180],"classical":[175],"techniques":[177],"significantly":[179],"cross-architecture":[181],"generalization,":[182],"requiring":[183],"processing":[184],"time":[185],"comparable":[186],"raw":[188],"DiT":[189],"ImageNet":[191],"(256$\\times$256)":[192],"with":[193],"4":[195],"GB":[196],"GPU":[198],"memory":[199],"usage.":[200]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-15T00:00:00"}