{"id":"https://openalex.org/W4417142499","doi":"https://doi.org/10.18653/v1/2026.acl-long.711","title":"To Think or Not to Think: The Hidden Cost of Meta-Training with Excessive CoT Examples","display_name":"To Think or Not to Think: The Hidden Cost of Meta-Training with Excessive CoT Examples","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4417142499","doi":"https://doi.org/10.18653/v1/2026.acl-long.711"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.acl-long.711","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.711","pdf_url":"https://aclanthology.org/2026.acl-long.711.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":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.711.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5081332799","display_name":"Vignesh Kothapalli","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vignesh Kothapalli","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5116358071","display_name":"Ata Fatahibaarzi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ata Fatahibaarzi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043413179","display_name":"Hamed Firooz","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hamed Firooz","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5046195860","display_name":"Maziar Sanjabi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Maziar Sanjabi","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":0.03419955,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"15619","last_page":"15644"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.5080000162124634,"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":0.5080000162124634,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.10949999839067459,"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":0.06019999831914902,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5616000294685364},{"id":"https://openalex.org/keywords/automated-reasoning","display_name":"Automated reasoning","score":0.3174999952316284},{"id":"https://openalex.org/keywords/case-based-reasoning","display_name":"Case-based reasoning","score":0.3084999918937683},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.2987000048160553},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.2702000141143799}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6467999815940857},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5616000294685364},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.490200012922287},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3433000147342682},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.3174999952316284},{"id":"https://openalex.org/C20162079","wikidata":"https://www.wikidata.org/wiki/Q1151406","display_name":"Case-based reasoning","level":2,"score":0.3084999918937683},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.2987000048160553},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.2847999930381775},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.2702000141143799},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.26649999618530273},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2563999891281128}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.711","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.711","pdf_url":"https://aclanthology.org/2026.acl-long.711.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"},{"id":"pmh:oai:arXiv.org:2512.05318","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2512.05318","pdf_url":"https://arxiv.org/pdf/2512.05318","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2512.05318","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2512.05318","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:10.18653/v1/2026.acl-long.711","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.711","pdf_url":"https://aclanthology.org/2026.acl-long.711.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":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4417142499.pdf","grobid_xml":"https://content.openalex.org/works/W4417142499.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Chain-of-thought":[0],"(CoT)":[1],"prompting":[2],"combined":[3],"with":[4,20],"few-shot":[5],"in-context":[6],"learning":[7],"(ICL)":[8],"has":[9],"unlocked":[10],"significant":[11],"reasoning":[12,58,146],"capabilities":[13],"in":[14,37,98,155],"large":[15],"language":[16],"models":[17],"(LLMs).However,":[18],"ICL":[19],"CoT":[21,61,76,94,125],"examples":[22,62,97,126],"is":[23,32,78,157],"ineffective":[24],"on":[25,113],"novel":[26,56,114],"tasks":[27,59,115,147],"when":[28,75,121],"the":[29,42,91,109,130],"pre-training":[30],"knowledge":[31],"insufficient.We":[33],"study":[34],"this":[35],"problem":[36],"a":[38,86],"controlled":[39],"setting":[40],"using":[41],"CoT-ICL":[43],"Lab":[44],"(Kothapalli":[45],"et":[46],"al.,":[47],"2025)":[48],"framework,":[49],"and":[50,95,148],"propose":[51,84],"meta-training":[52,99],"techniques":[53,135],"to":[54,89,118,139,153],"learn":[55],"abstract":[57],"in-context.Although":[60],"facilitate":[63],"reasoning,":[64],"we":[65,83],"noticed":[66],"that":[67,102],"their":[68],"excessive":[69],"inclusion":[70],"during":[71],"metatraining":[72],"degrades":[73],"performance":[74],"supervision":[77],"limited.To":[79],"mitigate":[80],"such":[81],"behavior,":[82],"CoT-Recipe,":[85],"formal":[87],"approach":[88],"modulate":[90],"mix":[92],"of":[93,111,133,151],"non-CoT":[96],"sequences.We":[100],"demonstrate":[101],"careful":[103],"modulation":[104],"via":[105],"CoT-Recipe":[106],"can":[107],"increase":[108],"accuracy":[110],"transformers":[112],"by":[116,136],"up":[117,152],"300%":[119],"even":[120],"there":[122],"are":[123],"no":[124],"available":[127,158],"in-context.We":[128],"confirm":[129],"broader":[131],"effectiveness":[132],"these":[134],"applying":[137],"them":[138],"pretrained":[140],"LLMs":[141],"(Qwen2.5":[142],"series)":[143],"for":[144],"symbolic":[145],"observing":[149],"gains":[150],"130%":[154],"accuracy.Code":[156],"at:":[159]},"counts_by_year":[],"updated_date":"2026-08-05T07:39:15.569665","created_date":"2025-12-09T00:00:00"}