{"id":"https://openalex.org/W7163349247","doi":"https://doi.org/10.48550/arxiv.2606.03054","title":"ToolGate: Token-Efficient Pre-Call Control for Tool-Augmented Vision-Language Agents","display_name":"ToolGate: Token-Efficient Pre-Call Control for Tool-Augmented Vision-Language Agents","publication_year":2026,"publication_date":"2026-06-02","ids":{"openalex":"https://openalex.org/W7163349247","doi":"https://doi.org/10.48550/arxiv.2606.03054"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.03054","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.03054","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":"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.03054","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137771980","display_name":"Anjie Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Anjie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137784421","display_name":"Yan Song","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Yan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137779836","display_name":"Zhixun Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Zhixun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137779102","display_name":"Ziqin Gong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gong, Ziqin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130223132","display_name":"Zhongwei Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Zhongwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137730239","display_name":"Jun Wang","orcid":"https://orcid.org/0000-0002-4957-6655"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jun","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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9617999792098999,"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":0.9617999792098999,"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/T10028","display_name":"Topic Modeling","score":0.004100000020116568,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.0032999999821186066,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.6812999844551086},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.6301000118255615},{"id":"https://openalex.org/keywords/controller","display_name":"Controller (irrigation)","score":0.592199981212616},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.5636000037193298},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.53329998254776},{"id":"https://openalex.org/keywords/simple","display_name":"Simple (philosophy)","score":0.3871999979019165}],"concepts":[{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.6812999844551086},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6452999711036682},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.6301000118255615},{"id":"https://openalex.org/C203479927","wikidata":"https://www.wikidata.org/wiki/Q5165939","display_name":"Controller (irrigation)","level":2,"score":0.592199981212616},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.5636000037193298},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.53329998254776},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.3871999979019165},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.36149999499320984},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3540000021457672},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2874999940395355},{"id":"https://openalex.org/C17500928","wikidata":"https://www.wikidata.org/wiki/Q959968","display_name":"Control system","level":2,"score":0.28139999508857727},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.27219998836517334}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.03054","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.03054","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.03054","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.03054","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":"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"},"sustainable_development_goals":[{"score":0.7098894119262695,"id":"https://metadata.un.org/sdg/1","display_name":"No poverty"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Tool-augmented":[0],"vision-language":[1],"agents":[2,149],"can":[3],"acquire":[4],"external":[5,94],"perceptual":[6,39,155],"evidence":[7],"through":[8],"OCR,":[9],"detection,":[10],"segmentation,":[11],"and":[12,23,69,103],"other":[13],"tools,":[14,156],"but":[15,157],"executing":[16],"every":[17],"proposed":[18],"tool":[19,40,164],"call":[20,44],"is":[21],"costly":[22],"sometimes":[24],"unnecessary.":[25],"We":[26,89],"study":[27],"the":[28,43,53,61,85,118],"pre-call":[29],"control":[30,161],"problem:":[31],"after":[32],"a":[33,38,92],"ReAct-style":[34],"VLM":[35,148],"agent":[36,63],"proposes":[37],"call,":[41],"should":[42],"be":[45],"executed,":[46],"or":[47],"skipped":[48],"before":[49],"its":[50],"output":[51],"enters":[52],"context?":[54],"Across":[55,107],"five":[56],"benchmarks,":[57],"we":[58],"find":[59],"that":[60,96,146],"baseline":[62,121],"exhibits":[64],"poor":[65],"local":[66],"selectivity:":[67],"helpful":[68],"harmful":[70],"calls":[71,81],"occur":[72],"at":[73],"similar":[74],"rates":[75],"(11.8%":[76],"vs.":[77],"9.9%),":[78],"while":[79,122],"most":[80],"do":[82],"not":[83,151],"change":[84],"immediate":[86],"forced-answer":[87],"prediction.":[88],"introduce":[90],"ToolGate,":[91],"lightweight":[93],"controller":[95],"predicts":[97],"execute/skip":[98],"decisions":[99],"from":[100,153,159],"trajectory":[101,131],"text":[102],"simple":[104],"structural":[105],"features.":[106],"two":[108],"Qwen3-VL":[109],"backbones,":[110],"ToolGate":[111],"reduces":[112],"token":[113],"cost":[114],"to":[115],"64-69%":[116],"of":[117],"unrestricted":[119],"ReAct":[120],"preserving":[123],"average":[124,138],"accuracy":[125,139],"in":[126],"cross-domain":[127],"settings.":[128],"With":[129],"matched-domain":[130],"training":[132],"on":[133],"Qwen3-VL-30B,":[134],"it":[135],"further":[136],"improves":[137],"by":[140],"1.65":[141],"points.":[142],"These":[143],"results":[144],"show":[145],"tool-augmented":[147],"benefit":[150],"only":[152],"better":[154],"also":[158],"explicit":[160],"over":[162],"when":[163],"outputs":[165],"are":[166],"worth":[167],"paying":[168],"for.":[169]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-04T00:00:00"}