{"id":"https://openalex.org/W7204251673","doi":"https://doi.org/10.48550/arxiv.2608.23635","title":"ToolRobustBench: Stage-Wise Perturbation Evaluation and Failure Diagnosis for Tool-Calling Agents","display_name":"ToolRobustBench: Stage-Wise Perturbation Evaluation and Failure Diagnosis for Tool-Calling Agents","publication_year":2026,"publication_date":"2026-08-23","ids":{"openalex":"https://openalex.org/W7204251673","doi":"https://doi.org/10.48550/arxiv.2608.23635"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.23635","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.23635","pdf_url":"https://arxiv.org/pdf/2608.23635","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.23635","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101256012","display_name":"Yishan ZHENG","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, YiShan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137917501","display_name":"Yuan Wu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Yuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5123469558","display_name":"Yi Chang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chang, Yi","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":true,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.6406000256538391,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.6406000256538391,"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":0.11320000141859055,"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/T10028","display_name":"Topic Modeling","score":0.04450000077486038,"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/tool-calling","display_name":"tool calling","score":0.8759999871253967},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.7310000061988831},{"id":"https://openalex.org/keywords/robustness-benchmark","display_name":"robustness benchmark","score":0.7639999985694885},{"id":"https://openalex.org/keywords/fault-diagnosis","display_name":"fault diagnosis","score":0.8169999718666077},{"id":"https://openalex.org/keywords/tool-selection","display_name":"tool selection","score":0.7250000238418579},{"id":"https://openalex.org/keywords/llm-based-agents","display_name":"LLM agents","score":0.5210000276565552}],"concepts":[{"id":"https://openalex.org/C177918212","wikidata":"https://www.wikidata.org/wiki/Q803623","display_name":"Perturbation (astronomy)","level":2,"score":0.6626999974250793},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.526199996471405},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.4934999942779541},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.3885999917984009},{"id":"https://openalex.org/C52146309","wikidata":"https://www.wikidata.org/wiki/Q7431116","display_name":"Schema (genetic algorithms)","level":2,"score":0.3447999954223633},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33880001306533813},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3165000081062317},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.28679999709129333},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2700999975204468},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.26350000500679016},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.26350000500679016},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2526000142097473}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.23635","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.23635","pdf_url":"https://arxiv.org/pdf/2608.23635","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.23635","doi":"https://doi.org/10.48550/arxiv.2608.23635","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.23635","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.23635","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.23635","pdf_url":"https://arxiv.org/pdf/2608.23635","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204251673.pdf","grobid_xml":"https://content.openalex.org/works/W7204251673.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7167627487","https://openalex.org/W7202365551","https://openalex.org/W7223575517","https://openalex.org/W7160993917","https://openalex.org/W7165870665","https://openalex.org/W7155074625","https://openalex.org/W4411119225","https://openalex.org/W7163681859","https://openalex.org/W7110894099","https://openalex.org/W7167691546"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"rely":[4],"on":[5,106],"tool":[6,93],"calling":[7],"as":[8],"a":[9,33,42,47,55,64,156],"fundamental":[10],"agent":[11,57],"capability,":[12],"enabling":[13],"them":[14],"to":[15,92],"invoke":[16],"external":[17],"systems":[18],"and":[19,69,86,101,120,129,158],"complete":[20],"tasks":[21],"beyond":[22,164],"text":[23],"generation.":[24],"However,":[25],"clean":[26,127,165],"end-to-end":[27],"(E2E)":[28],"success":[29],"cannot":[30],"identify":[31],"where":[32,54],"tool-use":[34,81],"failure":[35,143],"originates":[36],"or":[37],"how":[38],"it":[39],"propagates":[40],"through":[41],"call.":[43],"We":[44],"introduce":[45],"ToolRobustBench,":[46],"stage-wise":[48],"diagnostic":[49],"benchmark":[50,160],"for":[51,161],"tool-calling":[52,56,166],"agents,":[53],"is":[58],"an":[59],"LLM":[60],"system":[61],"that":[62,145],"selects":[63],"tool,":[65],"supplies":[66],"structured":[67],"arguments,":[68],"interprets":[70],"its":[71],"returned":[72],"feedback.":[73],"ToolRobustBench":[74,154],"aligns":[75],"four":[76],"perturbation":[77,118,135],"families":[78],"with":[79,133],"the":[80,136],"pipeline:":[82],"tool-interface,":[83],"user-intent,":[84],"tool-output/observation,":[85],"runtime-environment":[87],"perturbations.":[88],"It":[89],"attributes":[90],"failures":[91],"selection,":[94],"schema":[95],"grounding,":[96],"argument":[97],"binding,":[98],"tool-output/runtime-feedback":[99],"handling,":[100],"E2E":[102],"task":[103],"success.":[104],"Experiments":[105],"15,456":[107],"single-family":[108,151],"instances":[109],"across":[110],"7":[111],"models,":[112],"16":[113],"sampled":[114],"local":[115],"tools,":[116],"4":[117],"families,":[119],"14":[121],"subtypes":[122],"show":[123],"high":[124],"but":[125],"non-uniform":[126],"performance":[128],"substantial":[130],"robustness":[131,163],"degradation,":[132],"tool-output/observation":[134],"dominant":[137],"bottleneck.":[138],"Mixed-family":[139],"experiments":[140],"reveal":[141],"non-additive":[142],"patterns":[144],"are":[146],"not":[147],"explained":[148],"by":[149],"isolated":[150],"results.":[152],"Thus,":[153],"provides":[155],"deterministic":[157],"cascade-aware":[159],"diagnosing":[162],"accuracy;":[167]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-08-27T00:00:00"}