{"id":"https://openalex.org/W7164946119","doi":"https://doi.org/10.48550/arxiv.2606.17053","title":"Context-Aware RL for Agentic and Multimodal LLMs","display_name":"Context-Aware RL for Agentic and Multimodal LLMs","publication_year":2026,"publication_date":"2026-06-15","ids":{"openalex":"https://openalex.org/W7164946119","doi":"https://doi.org/10.48550/arxiv.2606.17053"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.17053","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.17053","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":"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.17053","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136134380","display_name":"Peiyang Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Peiyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138750808","display_name":"Bangzheng Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Bangzheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5103870133","display_name":"Liu S","orcid":"https://orcid.org/0000-0001-7255-0982"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Sijia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025205227","display_name":"Karthik Narasimhan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Narasimhan, Karthik R.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053980484","display_name":"Pramod Viswanath","orcid":"https://orcid.org/0000-0003-3171-8667"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Viswanath, Pramod","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5015619835","display_name":"Prateek Mittal","orcid":"https://orcid.org/0000-0002-4057-0118"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mittal, Prateek","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138694298","display_name":"Xingyu Fu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fu, Xingyu","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":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9786999821662903,"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":0.9786999821662903,"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":0.00570000009611249,"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":0.002899999963119626,"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/generative-grammar","display_name":"Generative grammar","score":0.5940999984741211},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.58160001039505},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.534500002861023},{"id":"https://openalex.org/keywords/trace","display_name":"TRACE (psycholinguistics)","score":0.5041000247001648},{"id":"https://openalex.org/keywords/coding","display_name":"Coding (social sciences)","score":0.4731000065803528},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.41600000858306885},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.38850000500679016},{"id":"https://openalex.org/keywords/similarity","display_name":"Similarity (geometry)","score":0.374099999666214}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7031000256538391},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.5940999984741211},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.58160001039505},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5422000288963318},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.534500002861023},{"id":"https://openalex.org/C75291252","wikidata":"https://www.wikidata.org/wiki/Q1315756","display_name":"TRACE (psycholinguistics)","level":2,"score":0.5041000247001648},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4952000081539154},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.4731000065803528},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.41600000858306885},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.38850000500679016},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3797000050544739},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.374099999666214},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.361299991607666},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.35199999809265137},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.33390000462532043},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.322299987077713},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3043999969959259},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.29750001430511475},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.2662999927997589},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.26570001244544983},{"id":"https://openalex.org/C155092808","wikidata":"https://www.wikidata.org/wiki/Q182557","display_name":"Computational linguistics","level":2,"score":0.265500009059906},{"id":"https://openalex.org/C76188268","wikidata":"https://www.wikidata.org/wiki/Q1783165","display_name":"Context effect","level":3,"score":0.2651999890804291},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.26260000467300415},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.2515000104904175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.17053","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.17053","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.17053","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.17053","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":"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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"often":[4],"fail":[5],"when":[6],"answering":[7,156],"requires":[8],"identifying":[9],"a":[10,18,25,29,33,42,72],"small":[11],"but":[12],"decisive":[13],"piece":[14],"of":[15,61,140,162,168],"evidence":[16],"within":[17],"long":[19],"or":[20,32],"complex":[21],"context,":[22],"such":[23],"as":[24,110,124,182],"single":[26],"line":[27],"in":[28,36,102],"tool":[30],"trace":[31],"subtle":[34],"detail":[35],"an":[37,56,74],"image.":[38],"We":[39,97],"propose":[40],"ContextRL,":[41],"context-aware":[43],"reinforcement":[44],"learning":[45],"(RL)":[46],"method":[47],"that":[48,88,167,176,194],"improves":[49],"long-horizon":[50,147],"reasoning":[51],"and":[52,76,81,133,149],"multimodal":[53,120],"performance":[54],"through":[55],"\\emph{indirect}":[57],"auxiliary":[58],"objective.":[59],"Instead":[60],"supervising":[62],"only":[63],"the":[64,69,86,90,160,163,178,195,199,206],"final":[65],"answer,":[66,75],"ContextRL":[67,136],"presents":[68],"model":[70],"with":[71],"query,":[73],"two":[77,103],"highly":[78],"similar":[79],"contexts,":[80,111,125],"rewards":[82],"it":[83],"for":[84,105,119],"selecting":[85],"context":[87,100],"supports":[89],"query--answer":[91],"pair,":[92],"thereby":[93],"encouraging":[94],"fine-grained":[95],"grounding.":[96],"construct":[98],"contrastive":[99,180,207],"data":[101,208],"domains:":[104],"coding":[106],"agents,":[107],"trajectories":[108],"serve":[109,123],"yielding":[112,126],"1k":[113],"pairs":[114,128],"built":[115,129],"via":[116,130],"condition":[117],"filtering;":[118],"reasoning,":[121],"images":[122],"7K":[127],"generative":[131],"editing":[132],"similarity":[134],"search.":[135],"achieves":[137],"average":[138],"gains":[139,196],"+2.2%":[141],"over":[142],"standard":[143,183],"GRPO":[144],"on":[145],"5":[146],"benchmarks,":[148],"+1.8%":[150],"across":[151],"12":[152],"diverse":[153],"visual":[154],"question":[155],"benchmarks.":[157],"To":[158],"disentangle":[159],"effect":[161],"proposed":[164,200],"objective":[165,202],"from":[166,198,205],"additional":[169],"data,":[170],"we":[171],"compare":[172],"against":[173],"data-augmentation":[174],"baselines":[175,187],"repurpose":[177],"same":[179],"contexts":[181],"query--context--answer":[184],"examples.":[185],"These":[186],"provide":[188],"little":[189],"to":[190],"no":[191],"improvement,":[192],"showing":[193],"arise":[197],"context-selection":[201],"rather":[203],"than":[204],"alone.":[209]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-17T00:00:00"}