{"id":"https://openalex.org/W4416034665","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.724","title":"From Remembering to Metacognition: Do Existing Benchmarks Accurately Evaluate LLMs?","display_name":"From Remembering to Metacognition: Do Existing Benchmarks Accurately Evaluate LLMs?","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034665","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.724"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.724","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.724","pdf_url":"https://aclanthology.org/2025.findings-emnlp.724.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":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-emnlp.724.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100354410","display_name":"Geng Zhang","orcid":"https://orcid.org/0000-0001-9038-4316"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Geng Zhang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Yizhou Ying","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yizhou Ying","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103193321","display_name":"Sihang Jiang","orcid":"https://orcid.org/0000-0002-0736-6457"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sihang Jiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075507821","display_name":"Jiaqing Liang","orcid":"https://orcid.org/0000-0003-0670-5602"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiaqing Liang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5033491042","display_name":"G. K. Yue","orcid":"https://orcid.org/0009-0000-3697-223X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guanglei Yue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077057825","display_name":"Yifei Fu","orcid":"https://orcid.org/0000-0001-9325-5114"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yifei Fu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103144672","display_name":"Hailin Hu","orcid":"https://orcid.org/0000-0003-3800-1511"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hailin Hu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5090455375","display_name":"Yanghua Xiao","orcid":"https://orcid.org/0000-0001-8403-9591"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yanghua Xiao","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":"13440","last_page":"13457"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10636","display_name":"Innovative Teaching and Learning Methods","score":0.1151999980211258,"subfield":{"id":"https://openalex.org/subfields/3204","display_name":"Developmental and Educational Psychology"},"field":{"id":"https://openalex.org/fields/32","display_name":"Psychology"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T10636","display_name":"Innovative Teaching and Learning Methods","score":0.1151999980211258,"subfield":{"id":"https://openalex.org/subfields/3204","display_name":"Developmental and Educational 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/T11902","display_name":"Intelligent Tutoring Systems and Adaptive Learning","score":0.06909999996423721,"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":0.04360000044107437,"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/process","display_name":"Process (computing)","score":0.27160000801086426},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.2655999958515167},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.2624000012874603},{"id":"https://openalex.org/keywords/identification","display_name":"Identification (biology)","score":0.25049999356269836},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.24619999527931213}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5934000015258789},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.34299999475479126},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.27160000801086426},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2655999958515167},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.2624000012874603},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.25049999356269836},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.24619999527931213},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.23600000143051147},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.23319999873638153},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.23190000653266907}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.724","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.724","pdf_url":"https://aclanthology.org/2025.findings-emnlp.724.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.724","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.724","pdf_url":"https://aclanthology.org/2025.findings-emnlp.724.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":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4416034665.pdf"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":null,"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-11-08T00:00:00"}
