{"id":"https://openalex.org/W4416034536","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.1344","title":"ToolHaystack: Stress-Testing Tool-Augmented Language Models in Realistic Long-Term Interactions","display_name":"ToolHaystack: Stress-Testing Tool-Augmented Language Models in Realistic Long-Term Interactions","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034536","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.1344"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.1344","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1344","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1344.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.1344.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Beong-woo Kwak","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Beong-woo Kwak","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5047557154","display_name":"Minju Kim","orcid":"https://orcid.org/0000-0001-5878-7350"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Minju Kim","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059758176","display_name":"Dongha Lim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dongha Lim","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Hyungjoo Chae","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hyungjoo Chae","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002352675","display_name":"Dong-Jin Kang","orcid":"https://orcid.org/0000-0001-8633-536X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dongjin Kang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100430153","display_name":"Sunghwan Kim","orcid":"https://orcid.org/0000-0002-1239-2350"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sunghwan Kim","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Dongil Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dongil Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5076900864","display_name":"Jinyoung Yeo","orcid":"https://orcid.org/0000-0003-3847-4917"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jinyoung Yeo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":8,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.1572522,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"24696","last_page":"24727"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.2903999984264374,"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":0.2903999984264374,"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/T12488","display_name":"Mental Health via Writing","score":0.1103999987244606,"subfield":{"id":"https://openalex.org/subfields/3207","display_name":"Social Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.08290000259876251,"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/natural-language","display_name":"Natural language","score":0.31380000710487366},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.3133000135421753},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.2761000096797943},{"id":"https://openalex.org/keywords/modeling-language","display_name":"Modeling language","score":0.26330000162124634},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.26030001044273376}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5950999855995178},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39739999175071716},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.35589998960494995},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.31380000710487366},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3133000135421753},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.2761000096797943},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.27140000462532043},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.26330000162124634},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.26030001044273376},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2554999887943268},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.2533000111579895},{"id":"https://openalex.org/C527412718","wikidata":"https://www.wikidata.org/wiki/Q855395","display_name":"Interpretation (philosophy)","level":2,"score":0.25119999051094055}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.1344","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1344","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1344.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.1344","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.1344","pdf_url":"https://aclanthology.org/2025.findings-emnlp.1344.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2257911995","display_name":null,"funder_award_id":"RS-2020-II201361","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"},{"id":"https://openalex.org/G4069530165","display_name":null,"funder_award_id":"2022-0-00077","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"},{"id":"https://openalex.org/G4700831490","display_name":null,"funder_award_id":"2022-","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"}],"funders":[{"id":"https://openalex.org/F4320321314","display_name":"Yonsei University","ror":"https://ror.org/01wjejq96"},{"id":"https://openalex.org/F4320328359","display_name":"Ministry of Science and ICT, South Korea","ror":"https://ror.org/01wpjm123"},{"id":"https://openalex.org/F4320335489","display_name":"Institute for Information and Communications Technology Promotion","ror":"https://ror.org/01g0hqq23"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034536.pdf","grobid_xml":"https://content.openalex.org/works/W4416034536.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2,73,92],"(LLMs)":[3],"have":[4],"demonstrated":[5],"strong":[6],"capabilities":[7,48],"in":[8,22,49,54,95,103,108],"using":[9],"external":[10],"tools":[11],"to":[12,83],"address":[13],"user":[14],"inquiries.However,":[15],"most":[16],"existing":[17],"evaluations":[18],"assume":[19],"tool":[20,46,116],"use":[21,47],"short":[23],"contexts,":[24],"offering":[25],"limited":[26],"insight":[27],"into":[28],"model":[29],"behavior":[30],"during":[31],"realistic":[32,62],"long-term":[33,50,110],"interactions.To":[34],"fill":[35],"this":[36,81],"gap,":[37],"we":[38,87],"introduce":[39],"TOOLHAYSTACK,":[40,104],"a":[41,65],"benchmark":[42,82],"for":[43],"testing":[44],"the":[45],"interactions.Each":[51],"test":[52],"instance":[53],"TOOLHAYSTACK":[55],"includes":[56],"multiple":[57],"tasks":[58],"execution":[59],"contexts":[60],"and":[61,76],"noise":[63],"within":[64],"continuous":[66],"conversation,":[67],"enabling":[68],"assessment":[69],"of":[70,126],"how":[71],"well":[72,94],"maintain":[74],"context":[75],"handle":[77],"various":[78],"disruptions.By":[79],"applying":[80],"14":[84],"state-of-the-art":[85],"LLMs,":[86],"find":[88],"that":[89],"while":[90],"current":[91],"perform":[93],"standard":[96],"multi-turn":[97],"settings,":[98],"they":[99],"often":[100],"significantly":[101],"struggle":[102],"highlighting":[105],"critical":[106],"gaps":[107],"their":[109],"robustness":[111],"not":[112],"revealed":[113],"by":[114],"previous":[115],"benchmarks.":[117],"1User:":[118],"It'll":[119],"be":[120],"30":[121],"people,":[122],"including":[123],"my":[124],"family":[125],"four.":[127]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-11-08T00:00:00"}