{"id":"https://openalex.org/W7153012613","doi":"https://doi.org/10.48550/arxiv.2604.07487","title":"CLEAR: Context Augmentation from Contrastive Learning of Experience via Agentic Reflection","display_name":"CLEAR: Context Augmentation from Contrastive Learning of Experience via Agentic Reflection","publication_year":2026,"publication_date":"2026-04-08","ids":{"openalex":"https://openalex.org/W7153012613","doi":"https://doi.org/10.48550/arxiv.2604.07487"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.07487","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07487","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.07487","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133361789","display_name":"Linbo Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Liu, Linbo","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5045718502","display_name":"Guande Wu","orcid":"https://orcid.org/0000-0002-9244-173X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Guande","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133357649","display_name":"Han Ding","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ding, Han","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133361172","display_name":"Yawei Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yawei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133386803","display_name":"Qiang Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Qiang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109474372","display_name":"Yuzhe Lu","orcid":"https://orcid.org/0009-0009-2882-2472"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Yuzhe","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133392077","display_name":"Zhichao Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Zhichao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133388874","display_name":"Huan Song","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Huan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133377212","display_name":"Panpan Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Panpan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133383736","display_name":"Lin Lee Cheong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cheong, Lin Lee","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":10,"corresponding_author_ids":["https://openalex.org/A5133361789"],"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.46650001406669617,"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"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.46650001406669617,"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/T10028","display_name":"Topic Modeling","score":0.10019999742507935,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.051500000059604645,"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":0.680400013923645},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6377999782562256},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5738000273704529},{"id":"https://openalex.org/keywords/reflection","display_name":"Reflection (computer programming)","score":0.5559999942779541},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.4771000146865845},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.4537999927997589},{"id":"https://openalex.org/keywords/context-model","display_name":"Context model","score":0.4424000084400177},{"id":"https://openalex.org/keywords/context-effect","display_name":"Context effect","score":0.42739999294281006}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7524999976158142},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.680400013923645},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6377999782562256},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5738000273704529},{"id":"https://openalex.org/C65682993","wikidata":"https://www.wikidata.org/wiki/Q1056451","display_name":"Reflection (computer programming)","level":2,"score":0.5559999942779541},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5292999744415283},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.4771000146865845},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4537999927997589},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.4424000084400177},{"id":"https://openalex.org/C76188268","wikidata":"https://www.wikidata.org/wiki/Q1783165","display_name":"Context effect","level":3,"score":0.42739999294281006},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.4180999994277954},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4007999897003174},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.39989998936653137},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3562999963760376},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.35040000081062317},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3312000036239624},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.32710000872612},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3264999985694885},{"id":"https://openalex.org/C59656382","wikidata":"https://www.wikidata.org/wiki/Q191536","display_name":"Conjunction (astronomy)","level":2,"score":0.30550000071525574},{"id":"https://openalex.org/C77967617","wikidata":"https://www.wikidata.org/wiki/Q4677561","display_name":"Active learning (machine learning)","level":2,"score":0.2867000102996826},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.27959999442100525},{"id":"https://openalex.org/C2777267654","wikidata":"https://www.wikidata.org/wiki/Q3519023","display_name":"Test (biology)","level":2,"score":0.27570000290870667},{"id":"https://openalex.org/C169903167","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Test set","level":2,"score":0.2671999931335449}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.07487","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07487","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.07487","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.07487","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":"article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/16","score":0.775044322013855,"display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"model":[2,7,118],"agents":[3],"rely":[4,21],"on":[5,22,58,168,192,203],"effective":[6],"context":[8,17,24,39,70,99,116,155],"to":[9,50,88,113,142,160,190,201],"obtain":[10],"task-relevant":[11],"information":[12],"for":[13,100],"decision-making.":[14],"Many":[15],"existing":[16],"engineering":[18],"approaches":[19],"primarily":[20],"the":[23,27,47,59,129,136,151,161,169],"generated":[25],"from":[26,40,150,188,199],"past":[28,41,93],"experience":[29],"and":[30,96,171,196],"retrieval":[31],"mechanisms":[32],"that":[33,156,177],"reuse":[34],"these":[35],"context.":[36],"However,":[37],"retrieved":[38],"tasks":[42],"must":[43],"be":[44],"adapted":[45],"by":[46,134],"execution":[48,94,138],"agent":[49,87],"fit":[51],"new":[52],"situations,":[53],"placing":[54],"additional":[55],"reasoning":[56],"burden":[57],"underlying":[60],"LLM.":[61],"To":[62],"address":[63],"this":[64],"limitation,":[65],"we":[66,121],"propose":[67],"a":[68,85,115,204],"generative":[69],"augmentation":[71,117],"framework":[72],"using":[73,125],"Contrastive":[74],"Learning":[75],"of":[76,206],"Experience":[77],"via":[78],"Agentic":[79],"Reflection":[80],"(CLEAR).":[81],"CLEAR":[82,178],"first":[83],"employs":[84],"reflection":[86],"perform":[89],"contrastive":[90],"analysis":[91],"over":[92],"trajectories":[95],"summarize":[97],"useful":[98],"each":[101],"observed":[102],"task.":[103,163],"These":[104],"summaries":[105],"are":[106],"then":[107],"used":[108],"as":[109],"supervised":[110],"fine-tuning":[111],"data":[112],"train":[114],"(CAM).":[119],"Then":[120],"further":[122],"optimize":[123],"CAM":[124,153],"reinforcement":[126],"learning,":[127],"where":[128],"reward":[130,198],"signal":[131],"is":[132,157,214],"obtained":[133],"running":[135],"task":[137,185],"agent.":[139,211],"By":[140],"learning":[141],"generate":[143],"task-specific":[144],"knowledge":[145,149],"rather":[146],"than":[147],"retrieve":[148],"past,":[152],"produces":[154],"better":[158],"tailored":[159],"current":[162],"We":[164],"conduct":[165],"comprehensive":[166],"evaluations":[167],"AppWorld":[170,193],"WebShop":[172],"benchmarks.":[173],"Experimental":[174],"results":[175],"show":[176],"consistently":[179],"outperforms":[180],"strong":[181],"baselines.":[182],"It":[183],"improves":[184],"completion":[186],"rate":[187],"72.62%":[189],"81.15%":[191],"test":[194],"set":[195],"averaged":[197],"0.68":[200],"0.74":[202],"subset":[205],"WebShop,":[207],"compared":[208],"with":[209],"baseline":[210],"Our":[212],"code":[213],"publicly":[215],"available":[216],"at":[217],"https://github.com/awslabs/CLEAR.":[218]},"counts_by_year":[],"updated_date":"2026-04-11T06:19:08.300824","created_date":"2026-04-11T00:00:00"}