{"id":"https://openalex.org/W7161246848","doi":"https://doi.org/10.48550/arxiv.2605.15102","title":"Improving Multi-turn Dialogue Consistency with Self-Recall Thinking","display_name":"Improving Multi-turn Dialogue Consistency with Self-Recall Thinking","publication_year":2026,"publication_date":"2026-05-14","ids":{"openalex":"https://openalex.org/W7161246848","doi":"https://doi.org/10.48550/arxiv.2605.15102"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.15102","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.15102","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,"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":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.15102","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136216672","display_name":"Renning Pang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pang, Renning","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136251389","display_name":"Tian Lan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lan, Tian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136231172","display_name":"Leyuan Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Leyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136242877","display_name":"Xiaoming Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Xiaoming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104504009","display_name":"Piao Tong","orcid":"https://orcid.org/0009-0008-7266-6828"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tong, Piao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136240619","display_name":"Xiaosong Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Xiaosong","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":false,"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":0.8087000250816345,"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":0.8087000250816345,"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":0.10400000214576721,"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/T10465","display_name":"Neurobiology of Language and Bilingualism","score":0.00930000003427267,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/recall","display_name":"Recall","score":0.7247999906539917},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5072000026702881},{"id":"https://openalex.org/keywords/perspective","display_name":"Perspective (graphical)","score":0.48910000920295715},{"id":"https://openalex.org/keywords/consistency","display_name":"Consistency (knowledge bases)","score":0.48399999737739563},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.4650999903678894},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.4602999985218048},{"id":"https://openalex.org/keywords/dependency","display_name":"Dependency (UML)","score":0.4431000053882599},{"id":"https://openalex.org/keywords/precision-and-recall","display_name":"Precision and recall","score":0.4081999957561493}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7513999938964844},{"id":"https://openalex.org/C100660578","wikidata":"https://www.wikidata.org/wiki/Q18733","display_name":"Recall","level":2,"score":0.7247999906539917},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.544700026512146},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5072000026702881},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.48910000920295715},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.48399999737739563},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.4650999903678894},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4602999985218048},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.4431000053882599},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.41850000619888306},{"id":"https://openalex.org/C81669768","wikidata":"https://www.wikidata.org/wiki/Q2359161","display_name":"Precision and recall","level":2,"score":0.4081999957561493},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3944000005722046},{"id":"https://openalex.org/C103057564","wikidata":"https://www.wikidata.org/wiki/Q4751139","display_name":"Analytic reasoning","level":3,"score":0.35350000858306885},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.35010001063346863},{"id":"https://openalex.org/C12186640","wikidata":"https://www.wikidata.org/wiki/Q6815743","display_name":"Memory model","level":3,"score":0.3253999948501587},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.3005000054836273},{"id":"https://openalex.org/C76679254","wikidata":"https://www.wikidata.org/wiki/Q5165163","display_name":"Context-dependent memory","level":4,"score":0.29339998960494995},{"id":"https://openalex.org/C143587482","wikidata":"https://www.wikidata.org/wiki/Q1543216","display_name":"Iterative and incremental development","level":2,"score":0.2913999855518341},{"id":"https://openalex.org/C2779458634","wikidata":"https://www.wikidata.org/wiki/Q24963715","display_name":"Debiasing","level":2,"score":0.2777000069618225},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.27320000529289246},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.2694000005722046},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2596000134944916}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.15102","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.15102","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,"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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.15102","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.15102","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,"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":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"model":[2,98],"(LLM)":[3],"based":[4],"multi-turn":[5,81],"dialogue":[6,38],"systems":[7],"often":[8],"struggle":[9],"to":[10,71,91,99,138,153],"track":[11],"dependencies":[12],"across":[13],"non-adjacent":[14],"turns,":[15],"undermining":[16],"both":[17],"consistency":[18],"and":[19,28,76,88,102,129,156,173,188,190],"scalability.":[20],"As":[21],"conversations":[22],"lengthen,":[23],"essential":[24],"information":[25],"becomes":[26],"sparse":[27,77],"is":[29],"buried":[30],"in":[31,80],"irrelevant":[32],"context,":[33],"while":[34],"processing":[35],"the":[36,97],"entire":[37],"history":[39],"incurs":[40],"severe":[41],"efficiency":[42],"bottlenecks.":[43],"Existing":[44],"solutions":[45],"either":[46],"rely":[47],"on":[48,162],"high":[49],"latency":[50,176,187],"external":[51,121],"memory":[52],"or":[53],"lose":[54],"fine-grained":[55],"details":[56],"through":[57],"iterative":[58],"summarization.":[59],"In":[60],"this":[61],"paper,":[62],"we":[63],"propose":[64],"Self-Recall":[65],"Thinking":[66],"(SRT),":[67],"a":[68,183],"framework":[69],"designed":[70],"address":[72],"long-range":[73],"contextual":[74],"dependency":[75],"informative":[78],"signals":[79],"dialogue.":[82],"SRT":[83,123,167],"identifies":[84],"helpful":[85],"historical":[86],"turns":[87],"uses":[89],"them":[90],"generate":[92],"contextually":[93],"appropriate":[94],"responses,":[95],"enabling":[96],"selectively":[100],"recall":[101,118,143,155],"reason":[103],"over":[104,179],"context":[105],"during":[106],"inference.":[107],"This":[108],"process":[109,114],"yields":[110],"an":[111],"endogenous":[112],"reasoning":[113,140,157,186],"that":[115,166],"integrates":[116],"interpretable":[117],"steps":[119],"without":[120],"modules.":[122],"incorporates:":[124],"(1)":[125],"Dependency":[126],"Construction:":[127],"Generating":[128],"converting":[130],"it":[131],"into":[132],"self-recall":[133],"chains;":[134],"(2)Capability":[135],"Initialization:":[136],"Training":[137],"enable":[139],"chains":[141],"with":[142],"tokens":[144],"capability;":[145],"(3)Reasoning":[146],"Improvement:":[147],"Refining":[148],"accuracy":[149],"via":[150],"verifiable":[151],"rewards":[152],"optimize":[154],"for":[158],"correct":[159],"answers.":[160],"Experiments":[161],"multiple":[163],"datasets":[164],"demonstrate":[165],"improves":[168],"F1":[169],"score":[170],"by":[171,177],"4.7%":[172],"reduces":[174],"end-to-end":[175],"14.7%":[178],"prior":[180],"methods,":[181],"achieving":[182],"balance":[184],"between":[185],"accuracy,":[189],"outperforming":[191],"state-of-the-art":[192],"baselines.":[193]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-05-16T00:00:00"}