{"id":"https://openalex.org/W4417116171","doi":"https://doi.org/10.18653/v1/2026.acl-long.152","title":"COMPASS: Enhancing Agent Long-Horizon Reasoning with Evolving Context","display_name":"COMPASS: Enhancing Agent Long-Horizon Reasoning with Evolving Context","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W4417116171","doi":"https://doi.org/10.18653/v1/2026.acl-long.152"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2026.acl-long.152","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.152","pdf_url":"https://aclanthology.org/2026.acl-long.152.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.acl-long.152.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5039490242","display_name":"Guoyang Wan","orcid":"https://orcid.org/0000-0002-6700-9056"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guangya Wan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045202393","display_name":"Mingyang Ling","orcid":"https://orcid.org/0009-0000-7377-8911"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mingyang Ling","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011112944","display_name":"Xiaoqi Ren","orcid":"https://orcid.org/0000-0002-1121-9046"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaoqi Ren","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025334096","display_name":"Rujun Han","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Rujun Han","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100359839","display_name":"Sheng Li","orcid":"https://orcid.org/0000-0003-1205-8632"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sheng Li","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5024743253","display_name":"Zizhao Zhang","orcid":"https://orcid.org/0000-0002-2395-4734"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zizhao Zhang","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":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.03373209,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"3360","last_page":"3380"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.23690000176429749,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.23690000176429749,"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":0.13750000298023224,"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"}},{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.11509999632835388,"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/context","display_name":"Context (archaeology)","score":0.6995999813079834},{"id":"https://openalex.org/keywords/bottleneck","display_name":"Bottleneck","score":0.5593000054359436},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.5203999876976013},{"id":"https://openalex.org/keywords/workflow","display_name":"Workflow","score":0.41190001368522644},{"id":"https://openalex.org/keywords/hallucinating","display_name":"Hallucinating","score":0.3833000063896179},{"id":"https://openalex.org/keywords/context-model","display_name":"Context model","score":0.3643999993801117}],"concepts":[{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6995999813079834},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6621000170707703},{"id":"https://openalex.org/C2780513914","wikidata":"https://www.wikidata.org/wiki/Q18210350","display_name":"Bottleneck","level":2,"score":0.5593000054359436},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.5203999876976013},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5185999870300293},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.41190001368522644},{"id":"https://openalex.org/C2911011789","wikidata":"https://www.wikidata.org/wiki/Q130741","display_name":"Hallucinating","level":2,"score":0.3833000063896179},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3702999949455261},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.3643999993801117},{"id":"https://openalex.org/C2778361833","wikidata":"https://www.wikidata.org/wiki/Q34735","display_name":"Compass","level":2,"score":0.3596000075340271},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.325300008058548},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.26750001311302185},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.2644999921321869},{"id":"https://openalex.org/C539667460","wikidata":"https://www.wikidata.org/wiki/Q2414942","display_name":"Management science","level":1,"score":0.26350000500679016},{"id":"https://openalex.org/C95423123","wikidata":"https://www.wikidata.org/wiki/Q622178","display_name":"Workgroup","level":2,"score":0.2619999945163727},{"id":"https://openalex.org/C114073186","wikidata":"https://www.wikidata.org/wiki/Q2631895","display_name":"Automated planning and scheduling","level":2,"score":0.2542000114917755}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.18653/v1/2026.acl-long.152","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.152","pdf_url":"https://aclanthology.org/2026.acl-long.152.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2510.08790","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.08790","pdf_url":"https://arxiv.org/pdf/2510.08790","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2510.08790","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.08790","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:10.18653/v1/2026.acl-long.152","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.acl-long.152","pdf_url":"https://aclanthology.org/2026.acl-long.152.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1565058763","display_name":null,"funder_award_id":"R01EB293388","funder_id":"https://openalex.org/F4320332161","funder_display_name":"National Institutes of Health"},{"id":"https://openalex.org/G5537453709","display_name":"Collaborative Research: III: Small: Physics Guided Graph Networks for Modeling Water Dynamics in Freshwater Ecosystems","funder_award_id":"2316306","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8191367516","display_name":"Collaborative Research: CPS: Medium: Multi-Scale Sensing and Control for Optimizing Crop Production and Energy Efficiency in Controlled-Environment Agriculture","funder_award_id":"2330215","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G8876996369","display_name":null,"funder_award_id":"N00014","funder_id":"https://openalex.org/F4320337345","funder_display_name":"Office of Naval Research"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320332161","display_name":"National Institutes of Health","ror":"https://ror.org/01cwqze88"},{"id":"https://openalex.org/F4320337345","display_name":"Office of Naval Research","ror":"https://ror.org/00rk2pe57"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4417116171.pdf","grobid_xml":"https://content.openalex.org/works/W4417116171.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Long-horizon":[0],"tasks":[1],"that":[2,73,90,99,111,147,159],"require":[3],"sustained":[4],"reasoning":[5,92,119],"and":[6,20,66,79,93,102,106,125,155],"multiple":[7],"tool":[8,94],"interactions":[9],"remain":[10],"challenging":[11,122],"for":[12,117,166],"LLM":[13],"agents:":[14],"small":[15],"errors":[16],"compound":[17],"across":[18],"steps,":[19],"even":[21],"state-of-the-art":[22],"models":[23,165],"often":[24],"hallucinate":[25],"or":[26,43,53],"lose":[27],"coherence.We":[28],"identify":[29],"context":[30,80,161],"management":[31,162],"as":[32],"the":[33],"central":[34],"bottleneck-extended":[35],"histories":[36],"cause":[37],"agents":[38],"to":[39,51,133,136,150,163],"overlook":[40],"critical":[41],"evidence":[42],"become":[44],"distracted":[45],"by":[46,131],"irrelevant":[47],"information,":[48],"thus":[49],"failing":[50],"replan":[52],"reflect":[54],"from":[55],"previous":[56],"mistakes.To":[57],"address":[58],"this,":[59],"we":[60],"propose":[61],"COMPASS":[62],"(Context-Organized":[63],"Multi-Agent":[64],"Planning":[65],"Strategy":[67],"System),":[68],"a":[69,87,97,108,143,156],"lightweight":[70],"hierarchical":[71],"framework":[72],"separates":[74],"tactical":[75],"execution,":[76],"strategic":[77,104],"oversight,":[78],"organization":[81],"into":[82],"three":[83,121],"specialized":[84],"components:":[85],"(1)":[86],"Main":[88],"Agent":[89],"performs":[91],"use,":[95],"(2)":[96],"Meta-Thinker":[98],"monitors":[100],"progress":[101,115],"issues":[103],"interventions,":[105],"(3)":[107],"Context":[109],"Manager":[110],"maintains":[112],"concise,":[113],"relevant":[114],"briefs":[116],"different":[118],"stages.Across":[120],"benchmarks-GAIA,":[123],"BrowseComp,":[124],"Humanity's":[126],"Last":[127],"Exam-COMPASS":[128],"improves":[129],"accuracy":[130],"up":[132],"20%":[134],"relative":[135],"both":[137],"single-and":[138],"multi-agent":[139],"baselines.We":[140],"further":[141],"introduce":[142],"test-time":[144],"scaling":[145],"extension":[146],"elevates":[148],"performance":[149],"match":[151],"established":[152],"DeepResearch":[153],"agents,":[154],"posttraining":[157],"pipeline":[158],"delegates":[160],"smaller":[164],"enhanced":[167],"efficiency.":[168]},"counts_by_year":[],"updated_date":"2026-08-06T08:24:18.245995","created_date":"2025-10-14T00:00:00"}