{"id":"https://openalex.org/W7204779959","doi":"https://doi.org/10.18653/2026.emnlp-main.299","title":"ContextPilot: Teaching Agents for Proactive Context Management via Fine-grained RL","display_name":"ContextPilot: Teaching Agents for Proactive Context Management via Fine-grained RL","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7204779959","doi":"https://doi.org/10.18653/2026.emnlp-main.299"},"language":"en","primary_location":{"id":"doi:10.18653/2026.emnlp-main.299","doi":"https://doi.org/10.18653/2026.emnlp-main.299","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.299","pdf_url":"https://aclanthology.org/2026.emnlp-main.299.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","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":"conference"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://aclanthology.org/2026.emnlp-main.299.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139942133","display_name":"Zhuoshi Pan","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]},{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhuoshi Pan","raw_affiliation_strings":["Tencent Youtu Lab","Tsinghua University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent Youtu Lab","institution_ids":["https://openalex.org/I2250653659"]},{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060614058","display_name":"Qizhi Pei","orcid":"https://orcid.org/0000-0002-7242-422X","observed_orcids":["https://orcid.org/0000-0002-7242-422X"]},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qizhi Pei","raw_affiliation_strings":["Shanghai AI Lab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai AI Lab","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090635327","display_name":"Junru Lu","orcid":"https://orcid.org/0009-0004-9588-8766","observed_orcids":["https://orcid.org/0009-0004-9588-8766"]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Junru Lu","raw_affiliation_strings":["Tencent Youtu Lab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent Youtu Lab","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148762919","display_name":"Honglin Lin","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4391012619","display_name":"Shanghai Artificial Intelligence Laboratory","ror":"https://ror.org/03wkvpx79","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4391012619"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Honglin Lin","raw_affiliation_strings":["Shanghai AI Lab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai AI Lab","institution_ids":["https://openalex.org/I4391012619"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148827921","display_name":"H. Vicky Zhao","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"H. Vicky Zhao","raw_affiliation_strings":["Tsinghua University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tsinghua University","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123875660","display_name":"Di Yin","orcid":"https://orcid.org/0009-0006-7246-3102","observed_orcids":["https://orcid.org/0009-0006-7246-3102"]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"di Yin","raw_affiliation_strings":["Tencent Youtu Lab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent Youtu Lab","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5140179585","display_name":"Xing Sun","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xing Sun","raw_affiliation_strings":["Tencent Youtu Lab"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Tencent Youtu Lab","institution_ids":["https://openalex.org/I2250653659"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.69110783,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"6886","last_page":"6903"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10286","display_name":"Information Retrieval and Search Behavior","score":0.3472000062465668,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.3472000062465668,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10028","display_name":"Topic Modeling","score":0.2533999979496002,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.19789999723434448,"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-models","display_name":"large language models","score":0.6230000257492065},{"id":"https://openalex.org/keywords/long-term-memory","display_name":"long-term memory","score":0.8159999847412109},{"id":"https://openalex.org/keywords/entropy-change","display_name":"entropy change","score":0.5619999766349792},{"id":"https://openalex.org/keywords/credit-assignment","display_name":"credit assignment","score":0.8040000200271606},{"id":"https://openalex.org/keywords/long-context-question-answering","display_name":"long-context question answering","score":0.7879999876022339},{"id":"https://openalex.org/keywords/deep-search","display_name":"deep search","score":0.8009999990463257}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8317999839782715},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5863000154495239},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5511999726295471},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.48660001158714294},{"id":"https://openalex.org/C2776285913","wikidata":"https://www.wikidata.org/wiki/Q5165183","display_name":"Context management","level":3,"score":0.41290000081062317},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.39640000462532043},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.34950000047683716},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3450999855995178},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.34389999508857727},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.3393000066280365},{"id":"https://openalex.org/C167981619","wikidata":"https://www.wikidata.org/wiki/Q1685498","display_name":"Cross entropy","level":3,"score":0.33880001306533813},{"id":"https://openalex.org/C2781368080","wikidata":"https://www.wikidata.org/wiki/Q501688","display_name":"Context awareness","level":3,"score":0.323199987411499},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.32199999690055847},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.310699999332428},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.29190000891685486},{"id":"https://openalex.org/C53833338","wikidata":"https://www.wikidata.org/wiki/Q1061424","display_name":"Context switch","level":2,"score":0.2890999913215637},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.2881999909877777},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.28110000491142273},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.2727000117301941},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.27250000834465027},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.2556999921798706},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.25440001487731934},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.2508000135421753}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/2026.emnlp-main.299","doi":"https://doi.org/10.18653/2026.emnlp-main.299","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.299","pdf_url":"https://aclanthology.org/2026.emnlp-main.299.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","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":"conference"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2608.28476","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.28476","pdf_url":"https://arxiv.org/pdf/2608.28476","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.2608.28476","doi":"https://doi.org/10.48550/arxiv.2608.28476","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.28476","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":"doi:10.18653/2026.emnlp-main.299","doi":"https://doi.org/10.18653/2026.emnlp-main.299","is_oa":true,"landing_page_url":"https://doi.org/10.18653/2026.emnlp-main.299","pdf_url":"https://aclanthology.org/2026.emnlp-main.299.pdf","source":{"id":"https://openalex.org/S4363608991","display_name":"Conference on Empirical Methods in Natural Language Processing (EMNLP)","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":"conference"},"license":"public-domain","license_id":"https://openalex.org/licenses/public-domain","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204779959.pdf","grobid_xml":"https://content.openalex.org/works/W7204779959.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4415270781","https://openalex.org/W7125261105","https://openalex.org/W7160502237","https://openalex.org/W7163005121","https://openalex.org/W7171536440","https://openalex.org/W7154426196","https://openalex.org/W7214794012","https://openalex.org/W7167033830","https://openalex.org/W7203718731","https://openalex.org/W7147424830"],"abstract_inverted_index":{"Long-horizon":[0],"agentic":[1,121],"tasks":[2,180],"require":[3],"large":[4],"language":[5],"models":[6,36,199],"(LLMs)":[7],"to":[8,25,37,57,100,152],"iteratively":[9],"retrieve,":[10],"integrate,":[11],"and":[12,60,70,89,132,149,161,177,200],"maintain":[13],"dispersed":[14],"information":[15],"across":[16,196],"multi-turn":[17],"interactions,":[18],"but":[19],"preserving":[20],"all":[21,101,166],"interaction":[22],"histories":[23],"leads":[24],"a":[26,53,114,188],"continuously":[27],"growing":[28],"working":[29,41,191],"context.Recent":[30],"proactive":[31,115],"context":[32,42,78,103,116,134,144,148,155],"management":[33,79,117],"methods":[34],"allow":[35],"edit":[38],"their":[39,83],"own":[40],"with":[43,62,128,187],"specialized":[44],"tools,":[45],"yet":[46],"they":[47],"still":[48],"face":[49],"three":[50],"key":[51],"limitations:(1)":[52],"limited":[54],"toolset":[55,127],"restricted":[56],"search,":[58],"deletion,":[59],"summarization,":[61],"no":[63],"support":[64],"for":[65,119,143,158],"global":[66],"planning,":[67,129],"long-term":[68,130],"memory,":[69,131],"adaptive":[71],"compression;":[72],"(2)":[73],"inefficient":[74],"exploration":[75],"that":[76,94,169,182],"treats":[77],"actions":[80,105,157],"uniformly":[81],"despite":[82],"heterogeneous":[84],"impacts":[85],"on":[86,174],"final":[87,97],"outcomes;":[88],"(3)":[90],"coarse-grained":[91],"credit":[92],"assignment":[93],"assigns":[95],"the":[96,126],"trajectorylevel":[98],"reward":[99],"intermediate":[102],"editing":[104],"during":[106],"RL.To":[107],"bridge":[108],"these":[109,172],"gaps,":[110],"we":[111],"introduce":[112],"ContextPilot,":[113],"framework":[118],"long-horizon":[120],"reasoning.Our":[122],"approach":[123],"systematically":[124],"augments":[125],"soft":[133],"offloading":[135],"tools.We":[136],"further":[137],"propose":[138],"an":[139],"RL":[140],"method":[141],"tailored":[142],"management,":[145],"which":[146],"uses":[147],"entropy":[150],"variation":[151],"identify":[153],"critical":[154],"control":[156],"branch":[159],"sampling":[160],"estimates":[162],"action-level":[163],"advantages":[164],"from":[165],"branched":[167],"trajectories":[168],"pass":[170],"through":[171],"actions.Experiments":[173],"long-context":[175],"QA":[176],"deep":[178],"search":[179],"show":[181],"ContextPilot":[183],"achieves":[184],"stronger":[185],"performance":[186],"more":[189],"compact":[190],"context,":[192],"outperforming":[193],"existing":[194],"baselines":[195],"various":[197],"base":[198],"benchmarks.Code":[201],"is":[202],"available":[203],"at":[204],"this":[205],"URL.":[206]},"counts_by_year":[],"updated_date":"2026-10-31T12:09:11.096373","created_date":"2026-09-01T00:00:00"}