{"id":"https://openalex.org/W6948475480","doi":"https://doi.org/10.48550/arxiv.2503.06692","title":"InftyThink: Breaking the Length Limits of Long-Context Reasoning in Large Language Models","display_name":"InftyThink: Breaking the Length Limits of Long-Context Reasoning in Large Language Models","publication_year":2025,"publication_date":"2025-03-09","ids":{"openalex":"https://openalex.org/W6948475480","doi":"https://doi.org/10.48550/arxiv.2503.06692"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2503.06692","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.06692","pdf_url":"https://arxiv.org/pdf/2503.06692","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2503.06692","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Yan, Yuchen","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Yuchen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Shen, Yongliang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shen, Yongliang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Liu, Yang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Jiang, Jin","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Jin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100717443","display_name":"Mengdi Zhang","orcid":"https://orcid.org/0000-0002-3239-4804","observed_orcids":["https://orcid.org/0000-0002-3239-4804"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Mengdi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Shao, Jian","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shao, Jian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Zhuang, Yueting","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhuang, Yueting","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":true,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.899399995803833,"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":0.899399995803833,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.03400000184774399,"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":0.023900000378489494,"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/long-context-reasoning","display_name":"long-context reasoning","score":0.9929999709129333},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.8859999775886536},{"id":"https://openalex.org/keywords/iterative-reasoning","display_name":"iterative reasoning","score":0.8669999837875366},{"id":"https://openalex.org/keywords/computational-scaling","display_name":"computational scaling","score":0.460999995470047},{"id":"https://openalex.org/keywords/mathematical-reasoning","display_name":"mathematical reasoning","score":0.44600000977516174},{"id":"https://openalex.org/keywords/reasoning-depth","display_name":"reasoning depth","score":0.3310000002384186}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7124999761581421},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5598999857902527},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.5042999982833862},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.4691999852657318},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4674000144004822},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.46380001306533813},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.46369999647140503},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.4456999897956848},{"id":"https://openalex.org/C28034677","wikidata":"https://www.wikidata.org/wiki/Q17092530","display_name":"Interleaving","level":2,"score":0.44119998812675476},{"id":"https://openalex.org/C83725634","wikidata":"https://www.wikidata.org/wiki/Q7268699","display_name":"Qualitative reasoning","level":2,"score":0.4359000027179718},{"id":"https://openalex.org/C143587482","wikidata":"https://www.wikidata.org/wiki/Q1543216","display_name":"Iterative and incremental development","level":2,"score":0.41679999232292175},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.41670000553131104},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.38850000500679016},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.3873000144958496},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.3822000026702881},{"id":"https://openalex.org/C159032336","wikidata":"https://www.wikidata.org/wiki/Q2488768","display_name":"Non-monotonic logic","level":2,"score":0.3752000033855438},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.37439998984336853},{"id":"https://openalex.org/C2779982483","wikidata":"https://www.wikidata.org/wiki/Q6094420","display_name":"Iterative refinement","level":2,"score":0.3531999886035919},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.33250001072883606},{"id":"https://openalex.org/C193221554","wikidata":"https://www.wikidata.org/wiki/Q5153664","display_name":"Commonsense reasoning","level":2,"score":0.30559998750686646},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3050000071525574},{"id":"https://openalex.org/C37335422","wikidata":"https://www.wikidata.org/wiki/Q6888134","display_name":"Model-based reasoning","level":3,"score":0.3001999855041504},{"id":"https://openalex.org/C103057564","wikidata":"https://www.wikidata.org/wiki/Q4751139","display_name":"Analytic reasoning","level":3,"score":0.2939000129699707},{"id":"https://openalex.org/C34388435","wikidata":"https://www.wikidata.org/wiki/Q2267362","display_name":"Bounded function","level":2,"score":0.2928999960422516},{"id":"https://openalex.org/C107848011","wikidata":"https://www.wikidata.org/wiki/Q4680756","display_name":"Adaptive reasoning","level":4,"score":0.27730000019073486},{"id":"https://openalex.org/C129844170","wikidata":"https://www.wikidata.org/wiki/Q41299","display_name":"Quadratic equation","level":2,"score":0.2587999999523163},{"id":"https://openalex.org/C127964446","wikidata":"https://www.wikidata.org/wiki/Q1092142","display_name":"Computational resource","level":3,"score":0.2581999897956848},{"id":"https://openalex.org/C110251889","wikidata":"https://www.wikidata.org/wiki/Q1569697","display_name":"Model checking","level":2,"score":0.2563000023365021},{"id":"https://openalex.org/C19754495","wikidata":"https://www.wikidata.org/wiki/Q3435924","display_name":"Computational problem","level":2,"score":0.2542000114917755},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.2535000145435333}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2503.06692","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.06692","pdf_url":"https://arxiv.org/pdf/2503.06692","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2503.06692","doi":"https://doi.org/10.48550/arxiv.2503.06692","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2503.06692","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:2503.06692","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2503.06692","pdf_url":"https://arxiv.org/pdf/2503.06692","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":null,"license_id":null,"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":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7126161078","https://openalex.org/W6891927978","https://openalex.org/W4416046347","https://openalex.org/W7148909843","https://openalex.org/W4407370164","https://openalex.org/W4417440832","https://openalex.org/W7083818123","https://openalex.org/W7160824387","https://openalex.org/W6910422195","https://openalex.org/W7162463770"],"abstract_inverted_index":{"Advanced":[0],"reasoning":[1,17,28,45,65,76,86,117,162,174],"in":[2],"large":[3],"language":[4],"models":[5],"has":[6],"achieved":[7],"remarkable":[8],"performance":[9,35],"on":[10],"challenging":[11],"tasks,":[12],"but":[13],"the":[14,49,158],"prevailing":[15],"long-context":[16,116],"paradigm":[18,61],"faces":[19],"critical":[20],"limitations:":[21],"quadratic":[22],"computational":[23,91,103,139,165],"scaling":[24,51],"with":[25,70,78,144],"sequence":[26],"length,":[27],"constrained":[29],"by":[30],"maximum":[31],"context":[32,39],"boundaries,":[33],"and":[34,152,164],"degradation":[36],"beyond":[37],"pre-training":[38],"windows.":[40],"Existing":[41],"approaches":[42],"primarily":[43],"compress":[44],"chains":[46],"without":[47,175],"addressing":[48],"fundamental":[50],"problem.":[52],"To":[53],"overcome":[54],"these":[55],"challenges,":[56],"we":[57,110],"introduce":[58],"InftyThink,":[59],"a":[60,95,112,168],"that":[62,100,135],"transforms":[63],"monolithic":[64],"into":[66,119,125],"an":[67],"iterative":[68,121],"process":[69],"intermediate":[71],"summarization.":[72],"By":[73],"interleaving":[74],"short":[75],"segments":[77],"concise":[79],"progress":[80],"summaries,":[81],"our":[82,120,136],"approach":[83,137,171],"enables":[84],"unbounded":[85],"depth":[87,163],"while":[88,141],"maintaining":[89],"bounded":[90],"costs.":[92],"This":[93],"creates":[94],"characteristic":[96],"sawtooth":[97],"memory":[98],"pattern":[99],"significantly":[101],"reduces":[102,138],"complexity":[104],"compared":[105],"to":[106,172],"traditional":[107],"approaches.":[108],"Furthermore,":[109],"develop":[111],"methodology":[113],"for":[114],"reconstructing":[115],"datasets":[118],"format,":[122],"transforming":[123],"OpenR1-Math":[124],"333K":[126],"training":[127],"instances.":[128],"Experiments":[129],"across":[130,149],"multiple":[131],"model":[132],"architectures":[133],"demonstrate":[134],"costs":[140],"improving":[142],"performance,":[143],"Qwen2.5-Math-7B":[145],"showing":[146],"3-11%":[147],"improvements":[148],"MATH500,":[150],"AIME24,":[151],"GPQA_diamond":[153],"benchmarks.":[154],"Our":[155],"work":[156],"challenges":[157],"assumed":[159],"trade-off":[160],"between":[161],"efficiency,":[166],"providing":[167],"more":[169],"scalable":[170],"complex":[173],"architectural":[176],"modifications.":[177]},"counts_by_year":[],"updated_date":"2026-10-14T16:39:01.820549","created_date":"2025-10-10T00:00:00"}