{"id":"https://openalex.org/W7168819865","doi":"https://doi.org/10.48448/0jjs-vw34","title":"AIPO: Adaptive Information Guided Token-Level Reinforcement Learning for Large Language Model Reasoning","display_name":"AIPO: Adaptive Information Guided Token-Level Reinforcement Learning for Large Language Model Reasoning","publication_year":2026,"publication_date":"2026-06-03","ids":{"openalex":"https://openalex.org/W7168819865","doi":"https://doi.org/10.48448/0jjs-vw34"},"language":"en","primary_location":{"id":"doi:10.48448/0jjs-vw34","doi":"https://doi.org/10.48448/0jjs-vw34","is_oa":true,"landing_page_url":"https://doi.org/10.48448/0jjs-vw34","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"},"volume":null,"type":"other","is_traditional":false,"indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48448/0jjs-vw34","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140982166","display_name":"Association for Computational Linguistics 2026","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Association for Computational Linguistics 2026","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5141028074","display_name":"Binchen .","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"., Binchen","raw_affiliation_strings":["University of Chinese Academy of Sciences"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Chinese Academy of Sciences","institution_ids":["https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5141022858","display_name":"Furui Liu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Furui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125742833","display_name":"Wenxi Liu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Wenxi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102236174","display_name":"H Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Huiyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139740566","display_name":"Hongfei Ye","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ye, Hongfei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5040809339","display_name":"Y Victoria Zhang","orcid":"https://orcid.org/0000-0003-4496-2650","observed_orcids":["https://orcid.org/0000-0003-4496-2650","https://orcid.org/0009-0004-2024-2434","https://orcid.org/0000-0002-1083-459X","https://orcid.org/0000-0002-2522-2027","https://orcid.org/0000-0003-1217-3876","https://orcid.org/0009-0001-5370-4381","https://orcid.org/0009-0004-4860-1910","https://orcid.org/0009-0007-1388-3737","https://orcid.org/0009-0006-3503-7593"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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":0.453900009393692,"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":0.453900009393692,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.26440000534057617,"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":0.09440000355243683,"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/large-language-model-reasoning","display_name":"large language model reasoning","score":0.7360000014305115},{"id":"https://openalex.org/keywords/reinforcement-learning-with-verifiable-rewards","display_name":"reinforcement learning with verifiable rewards","score":0.9089999794960022},{"id":"https://openalex.org/keywords/mutual-information-estimation","display_name":"mutual information estimation","score":0.8119999766349792},{"id":"https://openalex.org/keywords/hilbert-schmidt-independence-criterion","display_name":"Hilbert-Schmidt independence criterion","score":0.9089999794960022},{"id":"https://openalex.org/keywords/random-fourier-features","display_name":"random Fourier features","score":0.8320000171661377},{"id":"https://openalex.org/keywords/information-gain","display_name":"information gain","score":0.5600000023841858},{"id":"https://openalex.org/keywords/token-selection","display_name":"token selection","score":0.5899999737739563}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7818999886512756},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.736299991607666},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6715999841690063},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5565999746322632},{"id":"https://openalex.org/C35651441","wikidata":"https://www.wikidata.org/wiki/Q625303","display_name":"Independence (probability theory)","level":2,"score":0.5461000204086304},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.54339998960495},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.475600004196167},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.4156999886035919},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4146000146865845},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.38089999556541443},{"id":"https://openalex.org/C152139883","wikidata":"https://www.wikidata.org/wiki/Q252973","display_name":"Mutual information","level":2,"score":0.37290000915527344},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.2964000105857849},{"id":"https://openalex.org/C94922259","wikidata":"https://www.wikidata.org/wiki/Q33215","display_name":"Constructed language","level":2,"score":0.2847000062465668},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.27570000290870667},{"id":"https://openalex.org/C2776230583","wikidata":"https://www.wikidata.org/wiki/Q1322198","display_name":"Spoken language","level":2,"score":0.26190000772476196}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48448/0jjs-vw34","doi":"https://doi.org/10.48448/0jjs-vw34","is_oa":true,"landing_page_url":"https://doi.org/10.48448/0jjs-vw34","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"}],"best_oa_location":{"id":"doi:10.48448/0jjs-vw34","doi":"https://doi.org/10.48448/0jjs-vw34","is_oa":true,"landing_page_url":"https://doi.org/10.48448/0jjs-vw34","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7165614959","https://openalex.org/W7166849532","https://openalex.org/W4415025740","https://openalex.org/W7116110084","https://openalex.org/W4416781505","https://openalex.org/W7131402120","https://openalex.org/W4414898977","https://openalex.org/W4417438207","https://openalex.org/W4417299753","https://openalex.org/W4415234049"],"abstract_inverted_index":{"Reinforcement":[0],"Learning":[1],"with":[2],"Verifiable":[3],"Rewards":[4],"(RLVR)":[5],"improves":[6],"the":[7,46,80,91,125],"reasoning":[8],"capability":[9],"of":[10,82,90,115,127,137],"Large":[11],"Language":[12],"Models":[13],"(LLMs).":[14],"Current":[15],"RLVR":[16,109],"trains":[17],"LLMs":[18],"on":[19,40,45,64],"all":[20],"generated":[21],"tokens,":[22,67,116],"rather":[23],"than":[24],"exploring":[25],"which":[26,37],"tokens":[27,43],"actually":[28],"contribute":[29],"to":[30,55,104],"reasoning.":[31,139],"We":[32],"propose":[33],"AIPO(Adaptive--Information":[34],"Policy":[35,58],"Optimization),":[36],"focuses":[38],"updates":[39],"those":[41],"decisive":[42],"discovered":[44],"fly.":[47],"AIPO":[48,85,101],"estimates":[49],"each":[50],"hidden":[51],"state\u2019s":[52],"mutual":[53],"information":[54,73],"score":[56],"tokens.":[57],"gradients":[59],"are":[60],"then":[61],"computed":[62],"only":[63],"these":[65],"critical":[66],"using":[68],"an":[69],"advantage":[70],"that":[71],"blends":[72],"gain":[74],"and":[75,98,120,133],"verifiable":[76],"correctness.":[77],"To":[78],"improve":[79],"efficiency":[81,119],"mutual-information":[83],"estimation,":[84],"adopts":[86],"a":[87],"Random\u2013Fourier":[88],"approximation":[89],"Hilbert\u2013Schmidt":[92],"Independence":[93],"Criterion.":[94],"Across":[95],"five":[96],"math":[97],"science":[99],"benchmarks,":[100],"yields":[102],"up":[103],"+20%":[105],"accuracy":[106],"over":[107],"strong":[108],"baselines":[110],"while":[111],"updating":[112],"merely":[113],"10%":[114],"demonstrating":[117],"superior":[118],"effectiveness.":[121],"Our":[122],"findings":[123],"highlight":[124],"importance":[126],"information\u2013driven":[128],"token":[129],"selection":[130],"for":[131],"efficient":[132],"effective":[134],"reinforcement":[135],"learning":[136],"LLM":[138]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-07-17T00:00:00"}