{"id":"https://openalex.org/W7129576912","doi":"https://doi.org/10.48550/arxiv.2602.14635","title":"Alignment Adapter to Improve the Performance of Compressed Deep Learning Models","display_name":"Alignment Adapter to Improve the Performance of Compressed Deep Learning Models","publication_year":2026,"publication_date":"2026-02-16","ids":{"openalex":"https://openalex.org/W7129576912","doi":"https://doi.org/10.48550/arxiv.2602.14635"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.14635","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","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":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114114637","display_name":"Rohit Raj Rai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rai, Rohit Raj","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5099112809","display_name":"Abhishek Dhaka","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dhaka, Abhishek","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5126203568","display_name":"Amit Awekar","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Awekar, Amit","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/T10036","display_name":"Advanced Neural Network Applications","score":0.3693000078201294,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.3693000078201294,"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/T14347","display_name":"Big Data and Digital Economy","score":0.06599999964237213,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10273","display_name":"IoT and Edge/Fog Computing","score":0.03889999911189079,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/adapter","display_name":"Adapter (computing)","score":0.7702000141143799},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.6883999705314636},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.5199999809265137},{"id":"https://openalex.org/keywords/compression","display_name":"Compression (physics)","score":0.420199990272522},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.3716999888420105},{"id":"https://openalex.org/keywords/compressed-sensing","display_name":"Compressed sensing","score":0.3513000011444092}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7730000019073486},{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.7702000141143799},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.6883999705314636},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6468999981880188},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.5199999809265137},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.420199990272522},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3716999888420105},{"id":"https://openalex.org/C124851039","wikidata":"https://www.wikidata.org/wiki/Q2665459","display_name":"Compressed sensing","level":2,"score":0.3513000011444092},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3176000118255615},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.3165999948978424},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.3077999949455261},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.2921999990940094},{"id":"https://openalex.org/C127532173","wikidata":"https://www.wikidata.org/wiki/Q179904","display_name":"Hourglass","level":2,"score":0.26249998807907104},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.26030001044273376},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.25519999861717224},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2529999911785126}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.14635","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.14635","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.14635","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2602.14635","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","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":{"Compressed":[0],"Deep":[1],"Learning":[2],"(DL)":[3],"models":[4,105,121],"are":[5],"essential":[6],"for":[7,97],"deployment":[8],"in":[9,76,126],"resource-constrained":[10],"environments.":[11],"But":[12],"their":[13,18],"performance":[14,99,118],"often":[15],"lags":[16],"behind":[17],"large-scale":[19],"counterparts.":[20],"To":[21],"bridge":[22],"this":[23],"gap,":[24],"we":[25,111],"propose":[26],"Alignment":[27],"Adapter":[28],"(AlAd):":[29],"a":[30,40,80,84],"lightweight,":[31],"sliding-window-based":[32],"adapter.":[33],"It":[34],"aligns":[35],"the":[36,46,68,94,117],"token-level":[37,108],"embeddings":[38],"of":[39,45,119],"compressed":[41,86,95,120],"model":[42,96],"with":[43,93,122],"those":[44],"original":[47],"large":[48],"model.":[49],"AlAd":[50,72,92,114],"preserves":[51],"local":[52],"contextual":[53],"semantics,":[54],"enables":[55],"flexible":[56],"alignment":[57],"across":[58,106],"differing":[59],"dimensionalities":[60],"or":[61,88],"architectures,":[62],"and":[63,128],"is":[64],"entirely":[65],"agnostic":[66],"to":[67],"underlying":[69],"compression":[70],"method.":[71],"can":[73],"be":[74],"deployed":[75],"two":[77],"ways:":[78],"as":[79],"plug-and-play":[81],"module":[82],"over":[83],"frozen":[85],"model,":[87],"by":[89],"jointly":[90],"fine-tuning":[91],"further":[98],"gains.":[100],"Through":[101],"experiments":[102],"on":[103],"BERT-family":[104],"three":[107],"NLP":[109],"tasks,":[110],"demonstrate":[112],"that":[113],"significantly":[115],"boosts":[116],"only":[123],"marginal":[124],"overhead":[125],"size":[127],"latency.":[129]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-02-18T00:00:00"}