{"id":"https://openalex.org/W4402671969","doi":"https://doi.org/10.18653/v1/2024.acl-long.105","title":"Towards Faithful and Robust LLM Specialists for Evidence-Based Question-Answering","display_name":"Towards Faithful and Robust LLM Specialists for Evidence-Based Question-Answering","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W4402671969","doi":"https://doi.org/10.18653/v1/2024.acl-long.105"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2024.acl-long.105","doi":"https://doi.org/10.18653/v1/2024.acl-long.105","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.acl-long.105","pdf_url":"https://aclanthology.org/2024.acl-long.105.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"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 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2024.acl-long.105.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5008509425","display_name":"Tobias Schimanski","orcid":"https://orcid.org/0000-0002-3802-509X","observed_orcids":["https://orcid.org/0000-0002-3802-509X"]},"institutions":[{"id":"https://openalex.org/I202697423","display_name":"University of Zurich","ror":"https://ror.org/02crff812","country_code":"CH","type":"education","lineage":["https://openalex.org/I202697423"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Tobias Schimanski","raw_affiliation_strings":["University of Z\u00fcrich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Z\u00fcrich","institution_ids":["https://openalex.org/I202697423"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113207856","display_name":"Jingwei Ni","orcid":"https://orcid.org/0009-0008-8988-301X","observed_orcids":["https://orcid.org/0009-0008-8988-301X"]},"institutions":[{"id":"https://openalex.org/I202697423","display_name":"University of Zurich","ror":"https://ror.org/02crff812","country_code":"CH","type":"education","lineage":["https://openalex.org/I202697423"]},{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Jingwei Ni","raw_affiliation_strings":["ETH Z\u00fcrich","University of Z\u00fcrich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ETH Z\u00fcrich","institution_ids":["https://openalex.org/I35440088"]},{"raw_affiliation_string":"University of Z\u00fcrich","institution_ids":["https://openalex.org/I202697423"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5044596225","display_name":"Mathias Kraus","orcid":"https://orcid.org/0000-0002-2021-2743","observed_orcids":["https://orcid.org/0000-0002-2021-2743"]},"institutions":[{"id":"https://openalex.org/I60668342","display_name":"University of Regensburg","ror":"https://ror.org/01eezs655","country_code":"DE","type":"education","lineage":["https://openalex.org/I60668342"]}],"countries":["DE"],"is_corresponding":false,"raw_author_name":"Mathias Kraus","raw_affiliation_strings":["University of Regensburg"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Regensburg","institution_ids":["https://openalex.org/I60668342"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020377010","display_name":"Elliott Ash","orcid":"https://orcid.org/0000-0002-6817-7529","observed_orcids":["https://orcid.org/0000-0002-6817-7529"]},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Elliott Ash","raw_affiliation_strings":["ETH Z\u00fcrich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"ETH Z\u00fcrich","institution_ids":["https://openalex.org/I35440088"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5073309846","display_name":"Markus Leippold","orcid":"https://orcid.org/0000-0001-5983-2360","observed_orcids":["https://orcid.org/0000-0001-5983-2360"]},"institutions":[{"id":"https://openalex.org/I202697423","display_name":"University of Zurich","ror":"https://ror.org/02crff812","country_code":"CH","type":"education","lineage":["https://openalex.org/I202697423"]},{"id":"https://openalex.org/I54157616","display_name":"Swiss Finance Institute","ror":"https://ror.org/02cgn2a44","country_code":"CH","type":"education","lineage":["https://openalex.org/I54157616"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Markus Leippold","raw_affiliation_strings":["Swiss Finance Institute (SFI)","University of Z\u00fcrich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Swiss Finance Institute (SFI)","institution_ids":["https://openalex.org/I54157616"]},{"raw_affiliation_string":"University of Z\u00fcrich","institution_ids":["https://openalex.org/I202697423"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":5.8023,"has_fulltext":true,"cited_by_count":38,"citation_normalized_percentile":{"value":0.97078767,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":99,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"1913","last_page":"1931"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9502000212669373,"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":0.9502000212669373,"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.014000000432133675,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.011900000274181366,"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/large-language-models","display_name":"large language models","score":0.7379999756813049},{"id":"https://openalex.org/keywords/llm-fine-tuning","display_name":"LLM fine-tuning","score":0.7450000047683716},{"id":"https://openalex.org/keywords/source-quality","display_name":"source quality","score":0.8830000162124634},{"id":"https://openalex.org/keywords/synthetic-training-data","display_name":"synthetic training data","score":0.6589999794960022},{"id":"https://openalex.org/keywords/data-quality-filtering","display_name":"data quality filtering","score":0.8930000066757202},{"id":"https://openalex.org/keywords/robustness","display_name":"robustness","score":0.4959999918937683},{"id":"https://openalex.org/keywords/out-of-distribution-generalization","display_name":"out-of-distribution generalization","score":0.8569999933242798},{"id":"https://openalex.org/keywords/benchmark-dataset","display_name":"benchmark dataset","score":0.3190000057220459}],"concepts":[{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.6477668881416321},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5383539795875549},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.4040294885635376}],"mesh":[],"locations_count":10,"locations":[{"id":"doi:10.18653/v1/2024.acl-long.105","doi":"https://doi.org/10.18653/v1/2024.acl-long.105","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.acl-long.105","pdf_url":"https://aclanthology.org/2024.acl-long.105.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"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 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:www.zora.uzh.ch:270644","doi":null,"is_oa":true,"landing_page_url":"https://www.zora.uzh.ch/id/eprint/270644/1/2024.acl_long.105.pdf","pdf_url":"https://www.zora.uzh.ch/id/eprint/270644/1/2024.acl_long.105.pdf","source":{"id":"https://openalex.org/S4306401281","display_name":"Zurich Open Repository and Archive (University of Zurich)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I202697423","host_organization_name":"University of Zurich","host_organization_lineage":["https://openalex.org/I202697423"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"Schimanski, Tobias; Ni, Jingwei; Kraus, Mathias; Ash, Elliott; Leippold, Markus (2024). Towards Faithful and Robust LLM Specialists for Evidence-Based Question-Answering. In: The 62nd Annual Meeting of the Association for Computational Linguistics, Bangkok, Thailand, 11 August 2024 - 16 August 2024. Association for Computational Linguistics, 1913-1931.","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"pmh:oai:arXiv.org:2402.08277","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2402.08277","pdf_url":"https://arxiv.org/pdf/2402.08277","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:RePEc:chf:rpseri:rp2494","doi":null,"is_oa":false,"landing_page_url":"https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4728973","pdf_url":null,"source":{"id":"https://openalex.org/S4306401271","display_name":"RePEc: Research Papers in Economics","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I77793887","host_organization_name":"Federal Reserve Bank of St. Louis","host_organization_lineage":["https://openalex.org/I77793887"],"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":"preprint"},{"id":"pmh:oai:www.research-collection.ethz.ch:20.500.11850/717981","doi":null,"is_oa":true,"landing_page_url":"http://hdl.handle.net/20.500.11850/717981","pdf_url":null,"source":{"id":"https://openalex.org/S4306402302","display_name":"Repository for Publications and Research Data (ETH Zurich)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I35440088","host_organization_name":"ETH Zurich","host_organization_lineage":["https://openalex.org/I35440088"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"pmh:oai:www.research-collection.ethz.ch:20.500.11850/727648","doi":null,"is_oa":true,"landing_page_url":"http://hdl.handle.net/20.500.11850/727648","pdf_url":null,"source":{"id":"https://openalex.org/S4306402302","display_name":"Repository for Publications and Research Data (ETH Zurich)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I35440088","host_organization_name":"ETH Zurich","host_organization_lineage":["https://openalex.org/I35440088"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"info:eu-repo/semantics/conferenceObject"},{"id":"pmh:oai:www.zora.uzh.ch:20.500.14742/227662","doi":null,"is_oa":true,"landing_page_url":"https://www.zora.uzh.ch/handle/20.500.14742/227662","pdf_url":null,"source":{"id":"https://openalex.org/S4306401281","display_name":"Zurich Open Repository and Archive (University of Zurich)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I202697423","host_organization_name":"University of Zurich","host_organization_lineage":["https://openalex.org/I202697423"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Schimanski, T., Ni, J., Kraus, M., Ash, E., & Leippold, M. (2024). Towards Faithful and Robust LLM Specialists for Evidence-Based Question-Answering. Proceedings of the Annual Meeting of the Association for Computational Linguistics, 1913\u20131931. https://doi.org/10.18653/v1/2024.acl-long.105","raw_type":"Conference or Workshop Item"},{"id":"doi:10.48550/arxiv.2402.08277","doi":"https://doi.org/10.48550/arxiv.2402.08277","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2402.08277","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},{"id":"doi:10.3929/ethz-b-000717981","doi":"https://doi.org/10.3929/ethz-b-000717981","is_oa":true,"landing_page_url":"https://doi.org/10.3929/ethz-b-000717981","pdf_url":null,"source":{"id":"https://openalex.org/S7407051236","display_name":"ETH Z\u00fcrich Research Collection","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"article-journal"},{"id":"doi:10.5167/uzh-270644","doi":"https://doi.org/10.5167/uzh-270644","is_oa":true,"landing_page_url":"https://doi.org/10.5167/uzh-270644","pdf_url":null,"source":{"id":"https://openalex.org/S4306401281","display_name":"Zurich Open Repository and Archive (University of Zurich)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I202697423","host_organization_name":"University of Zurich","host_organization_lineage":["https://openalex.org/I202697423"],"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":"ConferencePaper"}],"best_oa_location":{"id":"doi:10.18653/v1/2024.acl-long.105","doi":"https://doi.org/10.18653/v1/2024.acl-long.105","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2024.acl-long.105","pdf_url":"https://aclanthology.org/2024.acl-long.105.pdf","source":{"id":"https://openalex.org/S4393915740","display_name":"Proceedings of the conference - Association for Computational Linguistics. Meeting","issn_l":"0736-587X","issn":["0736-587X"],"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":["norway-1"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"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 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G824780084","display_name":"How sustainable is sustainable finance? Impact evaluation and automated greenwashing detection.","funder_award_id":"207800","funder_id":"https://openalex.org/F4320320924","funder_display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320320924","display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","ror":"https://ror.org/00yjd3n13"},{"id":"https://openalex.org/F4320321942","display_name":"Hasler Stiftung","ror":"https://ror.org/04m3t9183"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4402671969.pdf","grobid_xml":"https://content.openalex.org/works/W4402671969.grobid-xml"},"referenced_works_count":5,"referenced_works":["https://openalex.org/W3015883388","https://openalex.org/W4387561528","https://openalex.org/W4387994989","https://openalex.org/W4388514555","https://openalex.org/W4389518954"],"related_works":["https://openalex.org/W7079579422","https://openalex.org/W4388684869","https://openalex.org/W7216338642","https://openalex.org/W4405783764","https://openalex.org/W4411953284","https://openalex.org/W4404579436","https://openalex.org/W4416526837","https://openalex.org/W4377865297","https://openalex.org/W4412888267","https://openalex.org/W4404782876"],"abstract_inverted_index":{"Advances":[0],"towards":[1],"more":[2,140],"faithful":[3],"and":[4,17,51,74,94],"traceable":[5],"answers":[6,28],"of":[7,44,108],"Large":[8],"Language":[9],"Models":[10],"(LLMs)":[11],"are":[12],"crucial":[13],"for":[14,70],"various":[15],"research":[16],"practical":[18],"endeavors.One":[19],"avenue":[20],"in":[21,42,143],"reaching":[22],"this":[23,32,60],"goal":[24],"is":[25],"basing":[26],"the":[27,46,54,106],"on":[29,116,121],"reliable":[30],"sources.However,":[31],"Evidence-Based":[33,145],"QA":[34],"has":[35],"proven":[36],"to":[37,66,104],"work":[38],"insufficiently":[39],"with":[40,83],"LLMs":[41,69],"terms":[43],"citing":[45],"correct":[47],"sources":[48,57],"(source":[49],"quality)":[50],"truthfully":[52],"representing":[53],"information":[55],"within":[56],"(answer":[58],"attributability).In":[59],"work,":[61],"we":[62,77,125],"systematically":[63],"investigate":[64],"how":[65],"robustly":[67],"fine-tune":[68],"better":[71],"source":[72],"quality":[73,86,137],"answer":[75],"attributability.Specifically,":[76],"introduce":[78,100],"a":[79],"data":[80,85,96,118,128],"generation":[81],"pipeline":[82],"automated":[84],"filters,":[87,138],"which":[88,130],"can":[89,131],"synthesize":[90],"diversified":[91],"high-quality":[92],"training":[93],"testing":[95],"at":[97],"scale.We":[98],"further":[99],"four":[101],"test":[102],"sets":[103],"benchmark":[105],"robustness":[107],"fine-tuned":[109],"specialist":[110],"models.Extensive":[111],"evaluation":[112],"shows":[113],"that":[114,127],"fine-tuning":[115],"synthetic":[117],"improves":[119],"performance":[120],"both":[122],"in-and":[123],"out-of-distribution.Furthermore,":[124],"show":[126],"quality,":[129],"be":[132],"drastically":[133],"improved":[134],"by":[135],"proposed":[136],"matters":[139],"than":[141],"quantity":[142],"improving":[144],"QA.":[146]},"counts_by_year":[{"year":2026,"cited_by_count":10},{"year":2025,"cited_by_count":17},{"year":2024,"cited_by_count":11}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2024-09-21T00:00:00"}