{"id":"https://openalex.org/W4416749302","doi":"https://doi.org/10.1109/iros60139.2025.11246298","title":"Task-Oriented Token Pruning for Efficient Object Detection and Segmentation","display_name":"Task-Oriented Token Pruning for Efficient Object Detection and Segmentation","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4416749302","doi":"https://doi.org/10.1109/iros60139.2025.11246298"},"language":null,"primary_location":{"id":"doi:10.1109/iros60139.2025.11246298","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros60139.2025.11246298","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","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/A5059667265","display_name":"Hao Liang","orcid":"https://orcid.org/0000-0003-4183-3423"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Liang","raw_affiliation_strings":["Chinese Academy of Sciences,State Key Laboratory of AI Safety, Institute of Computing Technology,Beijing,China,100190"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,State Key Laboratory of AI Safety, Institute of Computing Technology,Beijing,China,100190","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047682975","display_name":"Meina Kan","orcid":"https://orcid.org/0000-0001-9483-875X"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Meina Kan","raw_affiliation_strings":["Chinese Academy of Sciences,State Key Laboratory of AI Safety, Institute of Computing Technology,Beijing,China,100190"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,State Key Laboratory of AI Safety, Institute of Computing Technology,Beijing,China,100190","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5050297728","display_name":"Shiguang Shan","orcid":"https://orcid.org/0000-0002-8348-392X"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shiguang Shan","raw_affiliation_strings":["Chinese Academy of Sciences,State Key Laboratory of AI Safety, Institute of Computing Technology,Beijing,China,100190"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,State Key Laboratory of AI Safety, Institute of Computing Technology,Beijing,China,100190","institution_ids":["https://openalex.org/I4210090176"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5111681970","display_name":"Xilin Chen","orcid":"https://orcid.org/0009-0005-2633-8637"},"institutions":[{"id":"https://openalex.org/I4210090176","display_name":"Institute of Computing Technology","ror":"https://ror.org/0090r4d87","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210090176"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xilin Chen","raw_affiliation_strings":["Chinese Academy of Sciences,State Key Laboratory of AI Safety, Institute of Computing Technology,Beijing,China,100190"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,State Key Laboratory of AI Safety, Institute of Computing Technology,Beijing,China,100190","institution_ids":["https://openalex.org/I4210090176"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"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":0.32603738,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7826","last_page":"7833"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.7696999907493591,"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":0.7696999907493591,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.045499999076128006,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.02630000002682209,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.6621999740600586},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.49380001425743103},{"id":"https://openalex.org/keywords/object-detection","display_name":"Object detection","score":0.4916999936103821},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.43160000443458557},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.4203000068664551},{"id":"https://openalex.org/keywords/subnetwork","display_name":"Subnetwork","score":0.4180000126361847},{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.41029998660087585},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.3935000002384186},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.3871000111103058},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.36410000920295715}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8055999875068665},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.6621999740600586},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6496000289916992},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.49380001425743103},{"id":"https://openalex.org/C2776151529","wikidata":"https://www.wikidata.org/wiki/Q3045304","display_name":"Object detection","level":3,"score":0.4916999936103821},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.43160000443458557},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.4203000068664551},{"id":"https://openalex.org/C2780186347","wikidata":"https://www.wikidata.org/wiki/Q11414","display_name":"Subnetwork","level":2,"score":0.4180000126361847},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.41029998660087585},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3939000070095062},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.3935000002384186},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.3871000111103058},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.383899986743927},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.36410000920295715},{"id":"https://openalex.org/C138236772","wikidata":"https://www.wikidata.org/wiki/Q25098575","display_name":"Edge device","level":3,"score":0.3528999984264374},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.32989999651908875},{"id":"https://openalex.org/C64876066","wikidata":"https://www.wikidata.org/wiki/Q5141226","display_name":"Cognitive neuroscience of visual object recognition","level":3,"score":0.31119999289512634},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.30979999899864197},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.30970001220703125},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.30799999833106995},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.29840001463890076},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.296999990940094},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.2913999855518341},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.28870001435279846},{"id":"https://openalex.org/C5339829","wikidata":"https://www.wikidata.org/wiki/Q1425977","display_name":"Machine vision","level":2,"score":0.2883000075817108},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.28439998626708984},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.2831999957561493},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.2808000147342682},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.27379998564720154},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.266400009393692},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.26179999113082886},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.25780001282691956},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.25279998779296875},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.25060001015663147}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/iros60139.2025.11246298","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iros60139.2025.11246298","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W639708223","https://openalex.org/W1821462560","https://openalex.org/W1861492603","https://openalex.org/W2194775991","https://openalex.org/W2594519801","https://openalex.org/W2963140444","https://openalex.org/W2963150697","https://openalex.org/W2965862774","https://openalex.org/W3172942063","https://openalex.org/W3177196641","https://openalex.org/W3188427387","https://openalex.org/W4312290555","https://openalex.org/W4312872526","https://openalex.org/W4312910119","https://openalex.org/W4312980183","https://openalex.org/W4386071535","https://openalex.org/W4386071839","https://openalex.org/W4386076222","https://openalex.org/W4386566441","https://openalex.org/W4387949413","https://openalex.org/W4390873032","https://openalex.org/W4390874575","https://openalex.org/W4394597024","https://openalex.org/W4394625744","https://openalex.org/W4402716428","https://openalex.org/W4402753804","https://openalex.org/W4402959357"],"related_works":[],"abstract_inverted_index":{"Robots":[0],"rely":[1],"heavily":[2],"on":[3,50,88,144,225,230],"visual":[4],"perception":[5,18],"to":[6,94,105,119,166,199,222,241,253],"understand":[7],"and":[8,24,46,96,210],"interact":[9],"with":[10],"complex":[11],"environments.":[12],"To":[13,126],"support":[14,200],"this":[15,121,128],"capability,":[16],"modern":[17],"models":[19],"have":[20],"become":[21],"increasingly":[22],"large":[23],"powerful,":[25],"resulting":[26],"in":[27,36,67],"high":[28],"computational":[29,107],"costs":[30],"that":[31,138],"hinder":[32],"their":[33],"real-time":[34],"performance":[35,215],"robotic":[37,69,244],"applications.":[38],"Existing":[39,114],"acceleration":[40],"techniques,":[41],"such":[42],"as":[43],"model":[44],"pruning":[45,136,142,175],"token":[47,135,141,165],"pruning,":[48],"focus":[49,87],"reducing":[51],"architectural":[52],"or":[53],"parameter":[54],"redundancy":[55],"but":[56,92],"still":[57],"process":[58],"all":[59],"object":[60,79,236],"categories,":[61],"regardless":[62],"of":[63,78,163,216],"task":[64,238],"requirements.":[65],"However,":[66],"real-world":[68,243],"scenarios,":[70],"different":[71],"tasks":[72],"typically":[73],"require":[74],"only":[75],"a":[76,83,133,187],"subset":[77,203],"categories.":[80],"For":[81],"instance,":[82],"service":[84],"robot":[85],"may":[86],"kitchenware":[89],"while":[90,98],"cooking,":[91],"shift":[93],"furniture":[95],"obstacles":[97],"cleaning.":[99],"This":[100],"task-dependent":[101],"variation":[102],"creates":[103],"opportunities":[104],"reduce":[106],"cost":[108],"by":[109],"selectively":[110],"processing":[111,246],"relevant":[112],"information.":[113],"methods":[115],"are":[116,179],"not":[117],"designed":[118],"exploit":[120],"potential":[122],"for":[123,172],"task-specific":[124],"efficiency.":[125],"address":[127],"limitation,":[129],"we":[130],"propose":[131],"TaskTP,":[132],"task-oriented":[134],"method":[137],"dynamically":[139],"adjusts":[140],"based":[143],"the":[145,161,167,214,226,231],"target":[146],"category":[147,202],"set.":[148],"A":[149],"dynamic":[150],"gating":[151],"network":[152],"is":[153],"introduced":[154],"between":[155],"successive":[156],"Transformer":[157],"blocks,":[158],"which":[159],"evaluates":[160],"relevance":[162],"each":[164],"given":[168],"task.":[169],"TaskTP":[170,212],"allows":[171],"more":[173],"aggressive":[174],"when":[176],"fewer":[177],"categories":[178],"required,":[180],"optimizing":[181],"computation":[182],"without":[183,204],"sacrificing":[184],"performance.":[185],"After":[186],"task-agnostic":[188],"training":[189],"phase,":[190],"it":[191,207],"can":[192],"be":[193],"flexibly":[194],"configured":[195],"at":[196],"deployment":[197],"time":[198,247],"any":[201],"retraining,":[205],"making":[206],"both":[208],"efficient":[209],"versatile.":[211],"improves":[213],"Mask":[217],"R-CNN":[218],"from":[219,250],"31.4":[220],"fps":[221,224],"38.5":[223],"COCO":[227],"dataset.":[228],"Furthermore,":[229],"ScanNet":[232],"dataset,":[233],"where":[234],"an":[235],"search":[237],"was":[239,248],"defined":[240],"simulate":[242],"applications,":[245],"reduced":[249],"3197":[251],"ms":[252],"2437":[254],"ms,":[255],"demonstrating":[256],"significant":[257],"efficiency":[258],"gains.":[259]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-11-28T00:00:00"}