{"id":"https://openalex.org/W4402856762","doi":"https://doi.org/10.48550/arxiv.2407.14482","title":"ChatQA 2: Bridging the Gap to Proprietary LLMs in Long Context and RAG Capabilities","display_name":"ChatQA 2: Bridging the Gap to Proprietary LLMs in Long Context and RAG Capabilities","publication_year":2024,"publication_date":"2024-07-19","ids":{"openalex":"https://openalex.org/W4402856762","doi":"https://doi.org/10.48550/arxiv.2407.14482"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2407.14482","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2407.14482","pdf_url":"https://arxiv.org/pdf/2407.14482","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"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2407.14482","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","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":"Xu, Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113987537","display_name":"Wei Ping","orcid":"https://orcid.org/0000-0002-8852-6618","observed_orcids":["https://orcid.org/0000-0002-8852-6618"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ping, Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100548260","display_name":"Xianchao Wu","orcid":"https://orcid.org/0000-0002-6561-4789","observed_orcids":["https://orcid.org/0000-0002-6561-4789"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Xianchao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Xu, Chejian","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Chejian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","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":"Liu, Zihan","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":"Shoeybi, Mohammad","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":"Catanzaro, Bryan","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":44,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9829000234603882,"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":0.9829000234603882,"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":0.00570000009611249,"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"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.004999999888241291,"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/llama-3","display_name":"Llama 3","score":0.7580000162124634},{"id":"https://openalex.org/keywords/long-context-understanding","display_name":"long-context understanding","score":0.8180000185966492},{"id":"https://openalex.org/keywords/retrieval-augmented-generation","display_name":"retrieval-augmented generation","score":0.9850000143051147},{"id":"https://openalex.org/keywords/context-window-extension","display_name":"context window extension","score":0.8009999990463257},{"id":"https://openalex.org/keywords/instruction-tuning","display_name":"instruction tuning","score":0.8009999990463257},{"id":"https://openalex.org/keywords/open-source-llms","display_name":"open-source LLMs","score":0.7049999833106995},{"id":"https://openalex.org/keywords/long-context-llms","display_name":"long-context LLMs","score":0.5180000066757202}],"concepts":[{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.946151852607727},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.483578622341156},{"id":"https://openalex.org/C144133560","wikidata":"https://www.wikidata.org/wiki/Q4830453","display_name":"Business","level":0,"score":0.46316492557525635},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.23120474815368652},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.15438410639762878},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.14436250925064087},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2407.14482","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2407.14482","pdf_url":"https://arxiv.org/pdf/2407.14482","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.2407.14482","doi":"https://doi.org/10.48550/arxiv.2407.14482","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2407.14482","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:2407.14482","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2407.14482","pdf_url":"https://arxiv.org/pdf/2407.14482","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/W4402856762.pdf","grobid_xml":"https://content.openalex.org/works/W4402856762.grobid-xml"},"referenced_works_count":31,"referenced_works":["https://openalex.org/W2988841832","https://openalex.org/W4226227340","https://openalex.org/W4301243929","https://openalex.org/W4310923309","https://openalex.org/W4322718191","https://openalex.org/W4382603041","https://openalex.org/W4384918448","https://openalex.org/W4385681674","https://openalex.org/W4385890381","https://openalex.org/W4386435778","https://openalex.org/W4386978016","https://openalex.org/W4387559790","https://openalex.org/W4387596640","https://openalex.org/W4388183278","https://openalex.org/W4389519070","https://openalex.org/W4389520055","https://openalex.org/W4391556800","https://openalex.org/W4391876619","https://openalex.org/W4391901300","https://openalex.org/W4392617597","https://openalex.org/W4399115160","https://openalex.org/W4399795412","https://openalex.org/W4399836663","https://openalex.org/W4401042382","https://openalex.org/W4401306886","https://openalex.org/W4402671334","https://openalex.org/W4402671766","https://openalex.org/W4402672004","https://openalex.org/W4405595839","https://openalex.org/W4415798682","https://openalex.org/W7133203447"],"related_works":["https://openalex.org/W4404354805","https://openalex.org/W4404783788","https://openalex.org/W4387390366","https://openalex.org/W4404781754","https://openalex.org/W4407759431","https://openalex.org/W4402955078","https://openalex.org/W4399991263","https://openalex.org/W4407632460","https://openalex.org/W4414798800","https://openalex.org/W7101812775"],"abstract_inverted_index":{"In":[0],"this":[1],"work,":[2],"we":[3,165],"introduce":[4],"ChatQA":[5],"2,":[6],"an":[7],"Llama":[8],"3.0-based":[9],"model":[10,107,216],"with":[11,84],"a":[12,61,66,85,133,179,185],"128K":[13,81,211],"context":[14,32,74,135,141],"window,":[15,136],"designed":[16],"to":[17,44,51,71,80,90],"bridge":[18],"the":[19,73,92,105,128,138,159,168,198,215,221,225,227],"gap":[20],"between":[21,152],"open-source":[22,214],"LLMs":[23,50,173],"and":[24,34,47,97,116,155,205,210,220],"leading":[25],"proprietary":[26],"models":[27,202],"(e.g.,":[28,203],"GPT-4-Turbo-2024-04-09)":[29],"in":[30],"long":[31,140],"understanding":[33,99],"retrieval-augmented":[35],"generation":[36],"(RAG)":[37],"capabilities.":[38,100],"These":[39],"two":[40],"capabilities":[41],"are":[42],"complementary":[43],"each":[45],"other":[46],"essential":[48],"for":[49,224,226],"process":[52,89],"large":[53,186],"volumes":[54],"of":[55,76,170,182,188],"information":[56],"that":[57,104,167],"cannot":[58],"fit":[59],"into":[60],"single":[62],"prompt.":[63],"We":[64,147,213],"present":[65],"detailed":[67],"continued":[68],"training":[69,218],"recipe":[70],"extend":[72],"window":[75],"Llama3-70B-base":[77],"from":[78],"8K":[79],"tokens,":[82,123],"along":[83],"three-stage":[86],"instruction":[87],"tuning":[88],"enhance":[91],"model's":[93],"instruction-following,":[94],"RAG":[95,129,156,175,191],"performance,":[96],"long-context":[98,154,162,172,195,201],"Our":[101],"results":[102],"demonstrate":[103],"Llama3-ChatQA-2-70B":[106,204],"outperforms":[108,193],"most":[109],"existing":[110],"state-of-the-art":[111,161,200],"models,":[112],"including":[113],"GPT-4-Turbo-2024-04-09,":[114],"Qwen2-72B-Instruct,":[115],"Llama3.1-70B-Instruct,":[117],"on":[118,127,207],"ultra-long":[119],"tasks":[120],"beyond":[121],"100K":[122],"as":[124,126],"well":[125],"benchmark":[130],"using":[131,158,174,197],"only":[132],"4K":[134],"showing":[137],"strong":[139,171],"capability":[142],"across":[143],"varying":[144],"sequence":[145],"lengths.":[146],"further":[148],"provide":[149],"extensive":[150],"comparisons":[151],"direct":[153,194],"solutions":[157],"same":[160,199],"LLMs.":[163],"Interestingly,":[164],"find":[166],"performance":[169],"improves":[176],"when":[177],"retrieving":[178],"larger":[180],"number":[181],"chunks.":[183],"With":[184],"set":[187],"top-k":[189],"chunks,":[190],"consistently":[192],"solution":[196],"Qwen2-72B-Instruct)":[206],"both":[208],"32K":[209],"benchmarks.":[212],"weights,":[217],"data,":[219],"evaluation":[222],"setup":[223],"community:":[228],"https://chatqa2-project.github.io/":[229]},"counts_by_year":[{"year":2026,"cited_by_count":16},{"year":2025,"cited_by_count":25},{"year":2024,"cited_by_count":3}],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2024-09-26T00:00:00"}