{"id":"https://openalex.org/W7166859343","doi":"https://doi.org/10.18653/v1/2026.acl-long.354","title":"Revisiting the Reliability of Language Models in Instruction-Following","display_name":"Revisiting the Reliability of Language Models in Instruction-Following","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166859343","doi":"https://doi.org/10.18653/v1/2026.acl-long.354"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.354","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.354","pdf_url":"https://aclanthology.org/2026.acl-long.354.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":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.354.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5046878222","display_name":"Jianshuo Dong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jianshuo Dong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139847651","display_name":"Yutong Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yutong Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139781309","display_name":"Liu Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu Yan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139847587","display_name":"Zhenyu Zhong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhenyu Zhong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139758877","display_name":"Tao Wei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tao Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139771161","display_name":"Chao Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chao Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139823779","display_name":"Han Qiu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han Qiu","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.84790935,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"7784","last_page":"7812"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.2337999939918518,"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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.2337999939918518,"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/T13629","display_name":"Text Readability and Simplification","score":0.10369999706745148,"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":0.08569999784231186,"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/reliability","display_name":"Reliability (semiconductor)","score":0.5615000128746033},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.31790000200271606},{"id":"https://openalex.org/keywords/reliability-theory","display_name":"Reliability theory","score":0.3138999938964844},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.3073999881744385},{"id":"https://openalex.org/keywords/measure","display_name":"Measure (data warehouse)","score":0.29490000009536743}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.602400004863739},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.5615000128746033},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.36079999804496765},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.31790000200271606},{"id":"https://openalex.org/C201729545","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability theory","level":3,"score":0.3138999938964844},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3073999881744385},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.30730000138282776},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.29490000009536743},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.2883000075817108},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.27219998836517334},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.2556999921798706},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2533000111579895}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.354","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.354","pdf_url":"https://aclanthology.org/2026.acl-long.354.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":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.acl-long.354","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.354","pdf_url":"https://aclanthology.org/2026.acl-long.354.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":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[{"score":0.6371186971664429,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166859343.pdf","grobid_xml":"https://content.openalex.org/works/W7166859343.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Advanced":[0],"LLMs":[1],"have":[2],"achieved":[3],"near-ceiling":[4],"instruction-following":[5],"accuracy":[6],"on":[7],"benchmarks":[8],"such":[9],"as":[10,130],"IFEVAL.However,":[11],"these":[12],"impressive":[13],"scores":[14],"do":[15],"not":[16],"necessarily":[17],"translate":[18],"to":[19,110],"reliable":[20],"services":[21],"in":[22,102],"real-world":[23],"use,":[24],"where":[25],"users":[26],"often":[27],"vary":[28],"their":[29],"phrasing,":[30],"contextual":[31],"framing,":[32],"and":[33,67,90,120,140],"task":[34],"formulations.In":[35],"this":[36],"paper,":[37],"we":[38,61,82,94,117],"study":[39],"nuance-oriented":[40,103,128],"reliability:":[41],"whether":[42],"models":[43,98],"exhibit":[44,99],"consistent":[45],"competence":[46],"across":[47],"cousin":[48,75],"prompts":[49,76],"that":[50,72,96],"convey":[51],"analogous":[52],"user":[53],"intents":[54],"but":[55],"with":[56,112],"subtle":[57],"nuances.To":[58],"quantify":[59],"this,":[60,81],"introduce":[62],"a":[63,131],"new":[64],"metric,":[65],"reliable@k,":[66],"develop":[68],"an":[69],"automated":[70],"pipeline":[71],"generates":[73],"high-quality":[74],"via":[77],"data":[78],"augmentation.Building":[79],"upon":[80],"construct":[83],"IFE-VAL++":[84],"for":[85],"systematic":[86],"evaluation.Across":[87],"20":[88],"proprietary":[89],"26":[91],"open-source":[92],"LLMs,":[93],"find":[95],"current":[97],"substantial":[100],"insufficiency":[101],"reliability-their":[104],"performance":[105],"can":[106],"drop":[107],"by":[108],"up":[109],"61.8%":[111],"nuanced":[113],"prompt":[114],"modifications.What's":[115],"more,":[116],"characterize":[118],"it":[119],"explore":[121],"three":[122],"potential":[123],"improvement":[124],"recipes.Our":[125],"findings":[126],"highlight":[127],"reliability":[129],"crucial":[132],"yet":[133],"underexplored":[134],"next":[135],"step":[136],"toward":[137],"more":[138],"dependable":[139],"trustworthy":[141],"LLM":[142],"behavior.":[143]},"counts_by_year":[],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-07-02T00:00:00"}