{"id":"https://openalex.org/W4403791743","doi":"https://doi.org/10.1145/3664647.3680710","title":"Training-Free Feature Reconstruction with Sparse Optimization for Vision-Language Models","display_name":"Training-Free Feature Reconstruction with Sparse Optimization for Vision-Language Models","publication_year":2024,"publication_date":"2024-10-26","ids":{"openalex":"https://openalex.org/W4403791743","doi":"https://doi.org/10.1145/3664647.3680710"},"language":"en","primary_location":{"id":"doi:10.1145/3664647.3680710","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3664647.3680710","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5091022011","display_name":"Yi Zhang","orcid":"https://orcid.org/0000-0002-5831-0170"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]},{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yi Zhang","raw_affiliation_strings":["Harbin Institute of Technology & Southern University of Science and Technology, Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0002-5831-0170","affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology & Southern University of Science and Technology, Shenzhen, China","institution_ids":["https://openalex.org/I3045169105","https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055766208","display_name":"Ke Yu","orcid":"https://orcid.org/0009-0006-9687-6171"},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ke Yu","raw_affiliation_strings":["University of California, San Diego, San Diego, California, USA"],"raw_orcid":"https://orcid.org/0009-0006-9687-6171","affiliations":[{"raw_affiliation_string":"University of California, San Diego, San Diego, California, USA","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5013015879","display_name":"Angelica I. Avil\u00e9s-Rivero","orcid":"https://orcid.org/0000-0002-8878-0325"},"institutions":[{"id":"https://openalex.org/I241749","display_name":"University of Cambridge","ror":"https://ror.org/013meh722","country_code":"GB","type":"education","lineage":["https://openalex.org/I241749"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Angelica I. Aviles-Rivero","raw_affiliation_strings":["University of Cambridge, Cambridge, United Kingdom"],"raw_orcid":"https://orcid.org/0000-0002-8878-0325","affiliations":[{"raw_affiliation_string":"University of Cambridge, Cambridge, United Kingdom","institution_ids":["https://openalex.org/I241749"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077019861","display_name":"Jiyuan Jia","orcid":null},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiyuan Jia","raw_affiliation_strings":["Southern University of Science and Technology, Shenzhen, China"],"raw_orcid":"https://orcid.org/0009-0008-0519-6280","affiliations":[{"raw_affiliation_string":"Southern University of Science and Technology, Shenzhen, China","institution_ids":["https://openalex.org/I3045169105"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039193712","display_name":"Yushun Tang","orcid":"https://orcid.org/0000-0002-8350-7637"},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yushun Tang","raw_affiliation_strings":["Southern University of Science and Technology, Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0002-8350-7637","affiliations":[{"raw_affiliation_string":"Southern University of Science and Technology, Shenzhen, China","institution_ids":["https://openalex.org/I3045169105"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5045519742","display_name":"Zhihai He","orcid":"https://orcid.org/0000-0002-2647-8286"},"institutions":[{"id":"https://openalex.org/I3045169105","display_name":"Southern University of Science and Technology","ror":"https://ror.org/049tv2d57","country_code":"CN","type":"education","lineage":["https://openalex.org/I3045169105"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhihai He","raw_affiliation_strings":["Southern University of Science and Technology, Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0002-2647-8286","affiliations":[{"raw_affiliation_string":"Southern University of Science and Technology, Shenzhen, China","institution_ids":["https://openalex.org/I3045169105"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":6,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.17313407,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"4387","last_page":"4396"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9998999834060669,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9998999834060669,"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":0.9993000030517578,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9988999962806702,"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/computer-science","display_name":"Computer science","score":0.7435448169708252},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6997641921043396},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.6497069001197815},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.5030876994132996},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.45621269941329956},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.07188713550567627}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7435448169708252},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6997641921043396},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.6497069001197815},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.5030876994132996},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.45621269941329956},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.07188713550567627},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3664647.3680710","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3664647.3680710","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 32nd ACM International Conference on Multimedia","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":40,"referenced_works":["https://openalex.org/W12634471","https://openalex.org/W1928739709","https://openalex.org/W1977295328","https://openalex.org/W2017814585","https://openalex.org/W2047643928","https://openalex.org/W2108598243","https://openalex.org/W2130187411","https://openalex.org/W2136000821","https://openalex.org/W2138011018","https://openalex.org/W2155904486","https://openalex.org/W2194775991","https://openalex.org/W2533598788","https://openalex.org/W2962714319","https://openalex.org/W2963785020","https://openalex.org/W2964194231","https://openalex.org/W3012255272","https://openalex.org/W3044438666","https://openalex.org/W3109931228","https://openalex.org/W3168640669","https://openalex.org/W3173777717","https://openalex.org/W3182683290","https://openalex.org/W3185341429","https://openalex.org/W3198377975","https://openalex.org/W3205249428","https://openalex.org/W4212774754","https://openalex.org/W4226058394","https://openalex.org/W4229453513","https://openalex.org/W4304091651","https://openalex.org/W4304098884","https://openalex.org/W4312310776","https://openalex.org/W4312460555","https://openalex.org/W4312740349","https://openalex.org/W4312956471","https://openalex.org/W4313175608","https://openalex.org/W4382458283","https://openalex.org/W4386075788","https://openalex.org/W4386597150","https://openalex.org/W4390874269","https://openalex.org/W4393159766","https://openalex.org/W4403780517"],"related_works":["https://openalex.org/W2772917594","https://openalex.org/W2036807459","https://openalex.org/W2058170566","https://openalex.org/W2755342338","https://openalex.org/W2166024367","https://openalex.org/W3116076068","https://openalex.org/W2229312674","https://openalex.org/W2951359407","https://openalex.org/W2079911747","https://openalex.org/W1969923398"],"abstract_inverted_index":{"In":[0],"this":[1,86],"paper,":[2],"we":[3,89,118],"address":[4],"the":[5,31,47,51,54,72,76,104,120,124,137,141,154,159,164,172],"challenge":[6],"of":[7,53,63,80],"adapting":[8],"vision-language":[9],"models":[10],"(VLMs)":[11],"to":[12,28,122,135,152],"few-shot":[13,105,186,194],"image":[14,56,65,106,187],"recognition":[15,107,195],"in":[16,39,46,181,185],"a":[17,40,58,91,110],"training-free":[18,41],"manner.":[19],"We":[20,43,131],"observe":[21],"that":[22,197],"existing":[23,201],"methods":[24],"are":[25,70],"not":[26],"able":[27],"effectively":[29],"characterize":[30],"semantic":[32,48],"relationship":[33,173],"between":[34,174],"support":[35,64,127,143,175],"and":[36,60,74,113,126,176],"query":[37,55,125,138,177],"samples":[38],"setting.":[42],"recognize":[44],"that,":[45],"feature":[49,52,111,139,146],"space,":[50],"is":[57,149],"linear":[59],"sparse":[61,114,133],"combination":[62],"features":[66],"since":[67],"support-query":[68],"pairs":[69],"from":[71,140],"class":[73],"share":[75],"same":[77],"small":[78],"set":[79],"distinctive":[81],"visual":[82],"attributes.":[83],"Motivated":[84],"by":[85,163,204],"interesting":[87],"observation,":[88],"propose":[90],"novel":[92],"method":[93,199],"called":[94],"Training-free":[95],"Feature":[96],"ReConstruction":[97],"with":[98,158],"Sparse":[99],"optimization":[100,115,134],"(TaCo),":[101],"which":[102],"formulates":[103],"task":[108],"as":[109],"reconstruction":[112,147,155,167],"problem.":[116],"Specifically,":[117],"exploit":[119],"VLM":[121],"encode":[123],"images":[128],"into":[129],"features.":[130,144],"utilize":[132],"reconstruct":[136],"corresponding":[142],"The":[145],"error":[148],"then":[150],"used":[151],"define":[153],"similarity.":[156],"Coupled":[157],"text-image":[160],"similarity":[161,168],"provided":[162],"VLM,":[165],"our":[166,198],"analysis":[169],"accurately":[170],"characterizes":[171],"images.":[178],"This":[179],"results":[180,192],"significantly":[182],"improved":[183],"performance":[184],"recognition.":[188],"Our":[189],"extensive":[190],"experimental":[191],"on":[193],"demonstrate":[196],"outperforms":[200],"state-of-the-art":[202],"approaches":[203],"substantial":[205],"margins.":[206]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}