{"id":"https://openalex.org/W4403624735","doi":"https://doi.org/10.48550/arxiv.2409.06679","title":"E2LLM: Encoder Elongated Large Language Models for Long-Context Understanding and Reasoning","display_name":"E2LLM: Encoder Elongated Large Language Models for Long-Context Understanding and Reasoning","publication_year":2024,"publication_date":"2024-09-10","ids":{"openalex":"https://openalex.org/W4403624735","doi":"https://doi.org/10.48550/arxiv.2409.06679"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2409.06679","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2409.06679","pdf_url":"https://arxiv.org/pdf/2409.06679","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2409.06679","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5055616514","display_name":"Zihan Liao","orcid":"https://orcid.org/0000-0003-1574-5446"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Liao, Zihan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114860145","display_name":"Jun Wang","orcid":"https://orcid.org/0000-0003-2787-8932"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jun","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5062917240","display_name":"Hang Yu","orcid":"https://orcid.org/0000-0003-3444-9992"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Hang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005629203","display_name":"Lingxiao Wei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wei, Lingxiao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100368365","display_name":"Jianguo Li","orcid":"https://orcid.org/0000-0002-1147-4864"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Jianguo","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114860145","display_name":"Jun Wang","orcid":"https://orcid.org/0000-0003-2787-8932"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jun","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5112118564","display_name":"Wei Zhang","orcid":"https://orcid.org/0000-0001-7419-1100"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Wei","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5055616514"],"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":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":0.9086999893188477,"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":0.9086999893188477,"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/encoder","display_name":"Encoder","score":0.6687596440315247},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6376737356185913},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5686399340629578},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.46748679876327515},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.4231588840484619},{"id":"https://openalex.org/keywords/cognitive-science","display_name":"Cognitive science","score":0.371751606464386},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.36888137459754944},{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.2558780908584595},{"id":"https://openalex.org/keywords/geography","display_name":"Geography","score":0.11738625168800354},{"id":"https://openalex.org/keywords/philosophy","display_name":"Philosophy","score":0.11721748113632202}],"concepts":[{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.6687596440315247},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6376737356185913},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5686399340629578},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.46748679876327515},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.4231588840484619},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.371751606464386},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.36888137459754944},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.2558780908584595},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.11738625168800354},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.11721748113632202},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2409.06679","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2409.06679","pdf_url":"https://arxiv.org/pdf/2409.06679","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2409.06679","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2409.06679","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2409.06679","is_oa":true,"landing_page_url":"https://arxiv.org/abs/2409.06679","pdf_url":"https://arxiv.org/pdf/2409.06679","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4403624735.pdf","grobid_xml":"https://content.openalex.org/works/W4403624735.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4390516098","https://openalex.org/W2181948922","https://openalex.org/W2384362569","https://openalex.org/W2142795561","https://openalex.org/W4205302943","https://openalex.org/W2561132942","https://openalex.org/W3155418658","https://openalex.org/W4243199227","https://openalex.org/W2379948177","https://openalex.org/W2119949815"],"abstract_inverted_index":{"Processing":[0],"long":[1,63],"contexts":[2,64],"is":[3],"increasingly":[4],"important":[5],"for":[6,125],"Large":[7,50],"Language":[8,51],"Models":[9],"(LLMs)":[10],"in":[11,121],"tasks":[12],"like":[13],"multi-turn":[14],"dialogues,":[15],"code":[16],"generation,":[17],"and":[18,34,77,105,123,128],"document":[19,126],"summarization.":[20],"This":[21],"paper":[22],"addresses":[23],"the":[24,42,90,134],"challenges":[25],"of":[26,142],"achieving":[27],"high":[28],"long-context":[29,106],"performance,":[30],"low":[31],"computational":[32],"complexity,":[33],"compatibility":[35],"with":[36,81,93],"pretrained":[37,74],"models":[38,141],"--":[39],"collectively":[40],"termed":[41],"``impossible":[43],"triangle''.":[44],"We":[45],"introduce":[46],"E2LLM":[47,61,113],"(Encoder":[48],"Elongated":[49],"Models),":[52],"a":[53,73,82],"novel":[54],"approach":[55],"that":[56,112],"effectively":[57],"navigates":[58],"this":[59],"paradox.":[60],"divides":[62],"into":[65,69],"chunks,":[66],"compresses":[67],"each":[68],"soft":[70,95],"prompts":[71],"using":[72],"text":[75],"encoder,":[76],"aligns":[78],"these":[79,94],"representations":[80],"decoder-only":[83],"LLM":[84],"via":[85],"an":[86],"adapter.":[87],"To":[88],"enhance":[89],"LLM's":[91],"reasoning":[92],"prompts,":[96],"we":[97],"employ":[98],"two":[99],"training":[100],"objectives:":[101],"encoder":[102],"output":[103],"reconstruction":[104],"instruction":[107],"fine-tuning.":[108],"Extensive":[109],"experiments":[110],"reveal":[111],"not":[114],"only":[115],"outperforms":[116],"8":[117],"state-of-the-art":[118],"(SOTA)":[119],"methods":[120],"effectiveness":[122],"efficiency":[124],"summarization":[127],"question":[129],"answering,":[130],"but":[131],"also":[132],"achieves":[133],"best":[135],"performance":[136],"on":[137],"LongBench":[138],"v2":[139],"among":[140],"comparable":[143],"size.":[144]},"counts_by_year":[],"updated_date":"2026-04-17T18:11:37.981687","created_date":"2025-10-10T00:00:00"}