{"id":"https://openalex.org/W7161837884","doi":"https://doi.org/10.48550/arxiv.2605.19852","title":"Are Tools Always Beneficial? Learning to Invoke Tools Adaptively for Dual-Mode Multimodal LLM Reasoning","display_name":"Are Tools Always Beneficial? Learning to Invoke Tools Adaptively for Dual-Mode Multimodal LLM Reasoning","publication_year":2026,"publication_date":"2026-05-19","ids":{"openalex":"https://openalex.org/W7161837884","doi":"https://doi.org/10.48550/arxiv.2605.19852"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.19852","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.19852","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":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.2605.19852","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5104282179","display_name":"Qinghe Ma","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ma, Qinghe","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136570611","display_name":"Zhen Zhao","orcid":"https://orcid.org/0000-0002-7258-8865"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Zhen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136546042","display_name":"Yiming Wu","orcid":"https://orcid.org/0009-0008-9936-6737"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Yiming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136602375","display_name":"Jian Zhang","orcid":"https://orcid.org/0009-0004-9954-6050"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Jian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136555558","display_name":"Lei Bai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bai, Lei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136591986","display_name":"Yinghuan Shi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shi, Yinghuan","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/T10028","display_name":"Topic Modeling","score":0.4560999870300293,"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"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.4560999870300293,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.15240000188350677,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.07530000060796738,"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/focus","display_name":"Focus (optics)","score":0.5788000226020813},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5703999996185303},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.4708999991416931},{"id":"https://openalex.org/keywords/overhead","display_name":"Overhead (engineering)","score":0.4260999858379364},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.40369999408721924},{"id":"https://openalex.org/keywords/model-based-reasoning","display_name":"Model-based reasoning","score":0.3817000091075897},{"id":"https://openalex.org/keywords/visual-reasoning","display_name":"Visual reasoning","score":0.33640000224113464},{"id":"https://openalex.org/keywords/qualitative-reasoning","display_name":"Qualitative reasoning","score":0.3273000121116638}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7608000040054321},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6017000079154968},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.5788000226020813},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5703999996185303},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4708999991416931},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.4260999858379364},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.40369999408721924},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3880999982357025},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.3817000091075897},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.33640000224113464},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.33079999685287476},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.3273000121116638},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.31520000100135803},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.3131999969482422},{"id":"https://openalex.org/C103057564","wikidata":"https://www.wikidata.org/wiki/Q4751139","display_name":"Analytic reasoning","level":3,"score":0.3041999936103821},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.3034000098705292},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.30070000886917114},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.2838999927043915},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.2784999907016754},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.2599000036716461}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.19852","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.19852","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":"doi:10.48550/arxiv.2605.19852","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.19852","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":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"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],"reasoning":[1,11,54,94,116,126],"has":[2],"emerged":[3],"as":[4,47],"a":[5,68,85,114,148,161],"promising":[6],"direction":[7],"for":[8],"enhancing":[9],"the":[10,33,79,102,157],"capabilities":[12],"of":[13,35,81],"multimodal":[14],"large":[15],"language":[16],"models":[17,26],"(MLLMs).":[18],"However,":[19],"existing":[20,167],"studies":[21],"mainly":[22],"focus":[23],"on":[24,152,170],"enabling":[25],"to":[27,74,78,100,109,156],"perform":[28],"tool":[29,41],"invocation,":[30],"while":[31],"neglecting":[32],"necessity":[34],"invoking":[36],"tools.":[37],"We":[38],"argue":[39],"that":[40,70,139],"usage":[42],"is":[43,174],"not":[44],"always":[45],"beneficial,":[46],"redundant":[48],"or":[49],"inappropriate":[50],"invocations":[51],"largely":[52],"increase":[53],"overhead":[55],"and":[56,121,124,129,144,160],"even":[57],"mislead":[58],"model":[59,69,103],"predictions.":[60],"To":[61],"address":[62],"this":[63],"issue,":[64],"we":[65,89],"introduce":[66],"AutoTool,":[67],"adaptively":[71],"decides":[72],"whether":[73],"invoke":[75],"tools":[76],"according":[77],"characteristics":[80],"each":[82],"query.":[83],"Within":[84],"reinforcement":[86],"learning":[87],"framework,":[88],"design":[90],"an":[91],"explicit":[92],"dual-mode":[93],"strategy":[95],"with":[96],"mode-specific":[97],"reward":[98],"functions":[99],"guide":[101],"toward":[104,113],"producing":[105],"accurate":[106],"responses.":[107],"Moreover,":[108],"prevent":[110],"premature":[111],"bias":[112],"single":[115],"mode,":[117],"AutoTool":[118,140],"jointly":[119],"explores":[120],"balances":[122],"tool-assisted":[123],"text-centric":[125],"throughout":[127],"training,":[128],"promotes":[130],"free":[131],"exploration":[132],"in":[133,164],"later":[134],"stages.":[135],"Extensive":[136],"experiments":[137],"demonstrate":[138],"exhibits":[141],"outstanding":[142],"performance":[143],"high":[145],"efficiency,":[146],"yielding":[147],"21.8\\%":[149],"accuracy":[150],"gain":[151],"V*":[153],"benchmark":[154],"compared":[155],"base":[158],"model,":[159],"44.9\\%":[162],"improvement":[163],"efficiency":[165],"over":[166],"tool-augmented":[168],"methods":[169],"POPE":[171],"benchmark.":[172],"Code":[173],"available":[175],"at":[176],"https://github.com/MQinghe/AutoTool.":[177]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-21T00:00:00"}