{"id":"https://openalex.org/W4416375271","doi":"https://doi.org/10.48550/arxiv.2504.17892","title":"Token Sequence Compression for Efficient Multimodal Computing","display_name":"Token Sequence Compression for Efficient Multimodal Computing","publication_year":2025,"publication_date":"2025-04-24","ids":{"openalex":"https://openalex.org/W4416375271","doi":"https://doi.org/10.48550/arxiv.2504.17892"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2504.17892","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.17892","pdf_url":"https://arxiv.org/pdf/2504.17892","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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/2504.17892","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5117092446","display_name":"Yasmine Omri","orcid":"https://orcid.org/0009-0004-8444-2486","observed_orcids":["https://orcid.org/0009-0004-8444-2486"]},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Omri, Yasmine","raw_affiliation_strings":["Stanford University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120555402","display_name":"Parth Shroff","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Shroff, Parth","raw_affiliation_strings":["Stanford University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University","institution_ids":["https://openalex.org/I97018004"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5005762501","display_name":"Thierry Tambe","orcid":"https://orcid.org/0000-0002-6411-9620","observed_orcids":["https://orcid.org/0000-0002-6411-9620"]},"institutions":[{"id":"https://openalex.org/I97018004","display_name":"Stanford University","ror":"https://ror.org/00f54p054","country_code":"US","type":"education","lineage":["https://openalex.org/I97018004"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tambe, Thierry","raw_affiliation_strings":["Stanford University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Stanford University","institution_ids":["https://openalex.org/I97018004"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I97018004"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":9,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"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":1.0,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.0,"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/T11605","display_name":"Visual Attention and Saliency Detection","score":0.0,"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/vision-language-models","display_name":"vision-language models","score":0.7549999952316284},{"id":"https://openalex.org/keywords/multimodal-models","display_name":"multimodal models","score":0.6510000228881836},{"id":"https://openalex.org/keywords/vision-encoder","display_name":"vision encoder","score":0.6759999990463257},{"id":"https://openalex.org/keywords/token-selection","display_name":"token selection","score":0.8259999752044678},{"id":"https://openalex.org/keywords/token-merging","display_name":"token merging","score":0.9539999961853027},{"id":"https://openalex.org/keywords/cross-modal-attention","display_name":"cross-modal attention","score":0.9229999780654907}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.79830002784729},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7186999917030334},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.5741000175476074},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5497999787330627},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4916999936103821},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4885999858379364},{"id":"https://openalex.org/C2778869765","wikidata":"https://www.wikidata.org/wiki/Q6028363","display_name":"Inefficiency","level":2,"score":0.4528999924659729},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.4449999928474426},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.4090999960899353},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.39629998803138733},{"id":"https://openalex.org/C160086991","wikidata":"https://www.wikidata.org/wiki/Q5939193","display_name":"Human visual system model","level":3,"score":0.3652999997138977},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.3643999993801117},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.3635999858379364},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.3440999984741211},{"id":"https://openalex.org/C135641252","wikidata":"https://www.wikidata.org/wiki/Q738567","display_name":"Multimodal interaction","level":2,"score":0.3395000100135803},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.334199994802475},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.32919999957084656},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.32429999113082886},{"id":"https://openalex.org/C115067241","wikidata":"https://www.wikidata.org/wiki/Q1639854","display_name":"Token passing","level":3,"score":0.3100000023841858},{"id":"https://openalex.org/C2780385302","wikidata":"https://www.wikidata.org/wiki/Q367158","display_name":"Protocol (science)","level":3,"score":0.302700012922287},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.28780001401901245},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.27059999108314514},{"id":"https://openalex.org/C2778251979","wikidata":"https://www.wikidata.org/wiki/Q7936617","display_name":"Visual processing","level":3,"score":0.2630999982357025}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2504.17892","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.17892","pdf_url":"https://arxiv.org/pdf/2504.17892","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2504.17892","doi":"https://doi.org/10.48550/arxiv.2504.17892","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2504.17892","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2504.17892","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2504.17892","pdf_url":"https://arxiv.org/pdf/2504.17892","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416375271.pdf","grobid_xml":"https://content.openalex.org/works/W4416375271.grobid-xml"},"referenced_works_count":7,"referenced_works":["https://openalex.org/W3134144764","https://openalex.org/W4382142077","https://openalex.org/W4389523832","https://openalex.org/W4392222465","https://openalex.org/W4402727764","https://openalex.org/W4403081466","https://openalex.org/W4403346092"],"related_works":["https://openalex.org/W4415955360","https://openalex.org/W6967210562","https://openalex.org/W7155068464","https://openalex.org/W4409346467","https://openalex.org/W6979315904","https://openalex.org/W4416348134","https://openalex.org/W7155043590","https://openalex.org/W7140231574","https://openalex.org/W7122563326","https://openalex.org/W7164972282"],"abstract_inverted_index":{"The":[0],"exponential":[1],"growth":[2],"of":[3,56,114,134],"Large":[4],"Multimodal":[5],"Models":[6],"(LMMs)":[7],"has":[8],"driven":[9],"advancements":[10],"in":[11,34,82,101],"cross-modal":[12,119],"reasoning":[13],"but":[14],"at":[15,89],"significant":[16],"computational":[17],"costs.":[18],"In":[19,49,69],"this":[20,50],"work,":[21,51],"we":[22,52,71],"focus":[23],"on":[24,108],"visual":[25,57,115],"language":[26],"models.":[27],"We":[28,97],"highlight":[29],"the":[30,90,99,139],"redundancy":[31,100],"and":[32,38,60,66,85,94,105,132,137,144],"inefficiency":[33],"current":[35,102],"vision":[36,91,103],"encoders,":[37,104],"seek":[39],"to":[40],"construct":[41],"an":[42],"adaptive":[43],"compression":[44],"method":[45],"for":[46,141],"multimodal":[47,146],"data.":[48],"characterize":[53],"a":[54,125],"panoply":[55],"token":[58,76,83,116],"selection":[59,84,117],"merging":[61,88],"approaches":[62],"through":[63,118],"both":[64],"benchmarking":[65],"qualitative":[67],"analysis.":[68],"particular,":[70],"demonstrate":[72],"that":[73],"simple":[74],"cluster-level":[75],"aggregation":[77],"outperforms":[78],"prior":[79],"state-of-the-art":[80],"works":[81],"merging,":[86],"including":[87],"encoder":[92],"level":[93],"attention-based":[95],"approaches.":[96],"underline":[98],"shed":[106],"light":[107],"several":[109],"puzzling":[110],"trends":[111],"regarding":[112],"principles":[113],"attention":[120],"visualizations.":[121],"This":[122],"work":[123],"is":[124],"first":[126],"effort":[127],"towards":[128],"more":[129,142],"effective":[130],"encoding":[131],"processing":[133],"high-dimensional":[135],"data,":[136],"paves":[138],"way":[140],"scalable":[143],"sustainable":[145],"systems.":[147]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":4}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2025-10-10T00:00:00"}