{"id":"https://openalex.org/W4415798682","doi":"https://doi.org/10.52202/079017-0493","title":"ChatQA: Surpassing GPT-4 on Conversational QA and RAG","display_name":"ChatQA: Surpassing GPT-4 on Conversational QA and RAG","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W4415798682","doi":"https://doi.org/10.52202/079017-0493"},"language":"en","primary_location":{"id":"doi:10.52202/079017-0493","doi":"https://doi.org/10.52202/079017-0493","is_oa":false,"landing_page_url":"https://doi.org/10.52202/079017-0493","pdf_url":null,"source":{"id":"https://openalex.org/S4363606243","display_name":"neural information processing systems","issn_l":"1049-5258","issn":["1049-5258"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","norway-2"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Advances in Neural Information Processing Systems 37","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2401.10225","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100623107","display_name":"Zihan Liu","orcid":"https://orcid.org/0000-0002-5886-2792","observed_orcids":["https://orcid.org/0000-0002-5886-2792"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zihan Liu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Wei Ping","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei Ping","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Rajarshi Roy","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rajarshi Roy","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077532535","display_name":"Peng Xu","orcid":"https://orcid.org/0000-0003-3399-9722","observed_orcids":["https://orcid.org/0000-0003-3399-9722"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng Xu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5089891838","display_name":"Chankyu Lee","orcid":"https://orcid.org/0000-0002-6933-7459","observed_orcids":["https://orcid.org/0000-0002-6933-7459"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chankyu Lee","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072436307","display_name":"Mohammad Shoeybi","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mohammad Shoeybi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5066242985","display_name":"Bryan Catanzaro","orcid":"https://orcid.org/0000-0003-0034-7728","observed_orcids":["https://orcid.org/0000-0003-0034-7728"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bryan Catanzaro","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":3.176,"has_fulltext":true,"cited_by_count":11,"citation_normalized_percentile":{"value":0.93818731,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":99},"biblio":{"volume":null,"issue":null,"first_page":"15416","last_page":"15459"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.8773999810218811,"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":0.8773999810218811,"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/T12031","display_name":"Speech and dialogue systems","score":0.027899999171495438,"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":0.023600000888109207,"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/conversational-question-answering","display_name":"conversational question answering","score":0.953000009059906},{"id":"https://openalex.org/keywords/retrieval-augmented-generation","display_name":"retrieval-augmented generation","score":0.9440000057220459},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.8339999914169312},{"id":"https://openalex.org/keywords/gpt-4-comparison","display_name":"GPT-4 comparison","score":0.8209999799728394},{"id":"https://openalex.org/keywords/chatbots","display_name":"chatbots","score":0.43299999833106995}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5605999827384949},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3646000027656555},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3474000096321106},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.2930000126361847},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.28139999508857727},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.27869999408721924},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.2736999988555908},{"id":"https://openalex.org/C46312422","wikidata":"https://www.wikidata.org/wiki/Q11024","display_name":"Communication","level":1,"score":0.25270000100135803},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.24449999630451202},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2443999946117401}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.52202/079017-0493","doi":"https://doi.org/10.52202/079017-0493","is_oa":false,"landing_page_url":"https://doi.org/10.52202/079017-0493","pdf_url":null,"source":{"id":"https://openalex.org/S4363606243","display_name":"neural information processing systems","issn_l":"1049-5258","issn":["1049-5258"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["cwts-core","norway-2"],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Advances in Neural Information Processing Systems 37","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2401.10225","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2401.10225","pdf_url":"https://arxiv.org/pdf/2401.10225","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":"doi:10.48550/arxiv.2401.10225","doi":"https://doi.org/10.48550/arxiv.2401.10225","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2401.10225","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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2401.10225","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2401.10225","pdf_url":"https://arxiv.org/pdf/2401.10225","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"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4415798682.pdf","grobid_xml":"https://content.openalex.org/works/W4415798682.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4411549393","https://openalex.org/W4413967487","https://openalex.org/W4407214788","https://openalex.org/W7147412023","https://openalex.org/W7128621311","https://openalex.org/W4402345581","https://openalex.org/W4405205161","https://openalex.org/W4414523109","https://openalex.org/W7220949835","https://openalex.org/W7126162440"],"abstract_inverted_index":{"In":[0],"this":[1,145],"work,":[2],"we":[3,25,42,147],"introduce":[4,43],"ChatQA,":[5],"a":[6,27,44,98,138],"suite":[7],"of":[8,37,135],"models":[9],"that":[10,32],"outperform":[11,106],"GPT-4":[12],"on":[13,80,96,114,120],"retrieval-augmented":[14],"generation":[15],"(RAG)":[16],"and":[17,86,110,157],"conversational":[18,49],"question":[19],"answering":[20],"(QA).":[21],"To":[22,141],"enhance":[23],"generation,":[24],"propose":[26],"two-stage":[28],"instruction":[29,152],"tuning":[30,153],"method":[31],"significantly":[33],"boosts":[34],"the":[35,56,70,115,129,133,149,160],"performance":[36],"RAG.":[38],"For":[39],"effective":[40],"retrieval,":[41],"dense":[45],"retriever":[46,158],"optimized":[47],"for":[48,159],"QA,":[50,83],"which":[51,73],"yields":[52],"results":[53],"comparable":[54],"to":[55],"alternative":[57],"state-of-the-art":[58],"query":[59],"rewriting":[60],"models,":[61],"while":[62],"substantially":[63],"reducing":[64],"deployment":[65],"costs.":[66],"We":[67],"also":[68],"present":[69],"ChatRAG":[71,116,155],"Bench,":[72,117,156],"encompasses":[74],"ten":[75],"datasets":[76],"covering":[77],"comprehensive":[78],"evaluations":[79],"RAG,":[81],"table-related":[82],"arithmetic":[84],"calculations,":[85],"scenarios":[87],"involving":[88],"unanswerable":[89],"questions.":[90],"Our":[91],"ChatQA-1.0-70B":[92],"(score:":[93,108,112],"54.14),":[94],"built":[95],"Llama2,":[97],"weaker":[99],"foundation":[100],"model":[101,131,150],"than":[102],"GPT-4,":[103],"can":[104],"slightly":[105],"GPT-4-0613":[107],"53.90)":[109],"GPT-4-Turbo-2024-04-09":[111],"54.03)":[113],"without":[118],"relying":[119],"any":[121],"synthetic":[122],"data":[123],"from":[124],"OpenAI":[125],"GPT":[126],"models.":[127],"Notably,":[128],"Llama3-ChatQA-1.5-70B":[130],"surpasses":[132],"accuracy":[134],"GPT-4-Turbo-2024-04-09,":[136],"achieving":[137],"4.4%":[139],"improvement.":[140],"advance":[142],"research":[143],"in":[144],"field,":[146],"open-sourced":[148],"weights,":[151],"data,":[154],"community:":[161],"https://chatqa-project.github.io/.":[162]},"counts_by_year":[{"year":2026,"cited_by_count":9},{"year":2025,"cited_by_count":2}],"updated_date":"2026-10-21T09:56:26.102985","created_date":"2025-11-03T00:00:00"}