{"id":"https://openalex.org/W7160509947","doi":"https://doi.org/10.48550/arxiv.2605.04058","title":"MP-ISMoE: Mixed-Precision Interactive Side Mixture-of-Experts for Efficient Transfer Learning","display_name":"MP-ISMoE: Mixed-Precision Interactive Side Mixture-of-Experts for Efficient Transfer Learning","publication_year":2026,"publication_date":"2026-04-10","ids":{"openalex":"https://openalex.org/W7160509947","doi":"https://doi.org/10.48550/arxiv.2605.04058"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.04058","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.04058","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":null,"license_id":null,"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.2605.04058","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5135630661","display_name":"Yutong Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yutong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135549629","display_name":"Zimeng Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Zimeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135588926","display_name":"Shangcai Liao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liao, Shangcai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129739313","display_name":"Shujiang Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Shujiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135578362","display_name":"Jiaxin Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Jiaxin","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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.8079000115394592,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.8079000115394592,"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/T11448","display_name":"Face recognition and analysis","score":0.06040000170469284,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.030899999663233757,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/forgetting","display_name":"Forgetting","score":0.7202000021934509},{"id":"https://openalex.org/keywords/transfer-of-learning","display_name":"Transfer of learning","score":0.5720000267028809},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.5521000027656555},{"id":"https://openalex.org/keywords/boosting","display_name":"Boosting (machine learning)","score":0.534600019454956},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.49239999055862427},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.40709999203681946},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.3950999975204468},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.37880000472068787},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.36959999799728394}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7445999979972839},{"id":"https://openalex.org/C7149132","wikidata":"https://www.wikidata.org/wiki/Q1377840","display_name":"Forgetting","level":2,"score":0.7202000021934509},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.5720000267028809},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.5521000027656555},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.534600019454956},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5189999938011169},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.49239999055862427},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.40709999203681946},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.3950999975204468},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.37880000472068787},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.36959999799728394},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.3625999987125397},{"id":"https://openalex.org/C3454156","wikidata":"https://www.wikidata.org/wiki/Q1144241","display_name":"Side effect (computer science)","level":2,"score":0.35989999771118164},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3239000141620636},{"id":"https://openalex.org/C125112378","wikidata":"https://www.wikidata.org/wiki/Q176640","display_name":"Randomness","level":2,"score":0.3237000107765198},{"id":"https://openalex.org/C2780719617","wikidata":"https://www.wikidata.org/wiki/Q1030752","display_name":"Salient","level":2,"score":0.31450000405311584},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.2953000068664551},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2939000129699707},{"id":"https://openalex.org/C153258448","wikidata":"https://www.wikidata.org/wiki/Q1199743","display_name":"Gradient descent","level":3,"score":0.28060001134872437},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2775000035762787},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.26829999685287476},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.2678000032901764},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.2646999955177307},{"id":"https://openalex.org/C2781020372","wikidata":"https://www.wikidata.org/wiki/Q533093","display_name":"On the fly","level":2,"score":0.2614000141620636},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2565999925136566}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.04058","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.04058","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.04058","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.04058","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":null,"license_id":null,"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":{"Parameter-efficient":[0],"transfer":[1,36],"learning":[2,37,58],"(PETL)":[3],"has":[4],"emerged":[5],"as":[6],"a":[7,73,85],"pivotal":[8],"paradigm":[9],"for":[10],"adapting":[11],"pre-trained":[12],"foundation":[13],"models":[14],"to":[15,93,116,132,166],"downstream":[16],"tasks,":[17],"significantly":[18,64],"reducing":[19],"trainable":[20],"parameters":[21],"yet":[22],"suffering":[23],"from":[24,107,127,141],"substantial":[25],"memory":[26,54,105,124,175],"overhead":[27],"caused":[28],"by":[29,42,136],"gradient":[30,45],"backpropagation":[31],"during":[32],"fine-tuning.":[33],"While":[34],"memory-efficient":[35],"(METL)":[38],"circumvents":[39],"this":[40],"challenge":[41],"bypassing":[43],"backbone":[44],"computation":[46],"via":[47],"lightweight":[48],"small":[49],"side":[50,61,119],"networks,":[51,62],"its":[52],"stringent":[53],"constraint":[55],"severely":[56],"limits":[57],"capacity":[59],"of":[60],"thereby":[63],"compromising":[65],"performance.":[66,150],"To":[67],"address":[68],"these":[69],"limitations,":[70],"we":[71,82,109],"propose":[72,84],"novel":[74],"Mixed-Precision":[75],"Interactive":[76,112],"Side":[77,113],"Mixture-of-Experts":[78,114],"framework":[79],"(MP-ISMoE).":[80],"Specifically,":[81],"first":[83],"Gaussian":[86],"Noise":[87],"Perturbed":[88],"Iterative":[89],"Quantization":[90],"(GNP-IQ)":[91],"scheme":[92],"quantize":[94],"weights":[95],"into":[96],"lower-bits":[97],"while":[98,170],"effectively":[99],"decreasing":[100],"quantization":[101],"errors.":[102],"By":[103],"leveraging":[104],"conserved":[106],"GNP-IQ,":[108],"subsequently":[110],"employ":[111],"(ISMoE)":[115],"scaling":[117],"up":[118],"networks":[120],"without":[121],"sacrificing":[122],"overall":[123],"efficiency.":[125,176],"Different":[126],"conventional":[128],"mixture-of-experts,":[129],"ISMoE":[130],"learns":[131],"select":[133],"optimal":[134],"experts":[135],"interacting":[137],"with":[138],"salient":[139],"features":[140],"frozen":[142],"backbones,":[143],"thus":[144],"suppressing":[145],"knowledge":[146],"forgetting":[147],"and":[148,156,174],"boosting":[149],"Extensive":[151],"experiments":[152],"across":[153],"diverse":[154],"vision-language":[155],"language-only":[157],"tasks":[158],"demonstrate":[159],"that":[160],"MP-ISMoE":[161],"remarkably":[162],"promotes":[163],"accuracy":[164],"compared":[165],"state-of-the-art":[167],"METL":[168],"approaches,":[169],"maintaining":[171],"comparable":[172],"parameter":[173]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-05-08T00:00:00"}