{"id":"https://openalex.org/W7166019693","doi":"https://doi.org/10.48550/arxiv.2606.26918","title":"Diagnosing Task Insensitivity in Language Agents","display_name":"Diagnosing Task Insensitivity in Language Agents","publication_year":2026,"publication_date":"2026-06-25","ids":{"openalex":"https://openalex.org/W7166019693","doi":"https://doi.org/10.48550/arxiv.2606.26918"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2606.26918","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2606.26918","pdf_url":"https://arxiv.org/pdf/2606.26918","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2606.26918","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139441778","display_name":"Jingyu Liu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I78988378","display_name":"Renmin University of China","ror":"https://ror.org/041pakw92","country_code":"CN","type":"education","lineage":["https://openalex.org/I78988378"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liu, Jingyu","raw_affiliation_strings":["Gaoling School of Artificial Intelligence Renmin University of China , Beijing , China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Gaoling School of Artificial Intelligence Renmin University of China , Beijing , China","institution_ids":["https://openalex.org/I78988378"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114453024","display_name":"Jiaen Lin","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Jiaen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129751422","display_name":"xiaopeng Wu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wu, Xiaopeng","raw_affiliation_strings":["Taobao & Tmall Group of Alibaba"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Taobao & Tmall Group of Alibaba","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5159126067","display_name":"Kehan Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen, Kehan","raw_affiliation_strings":["Taobao & Tmall Group of Alibaba"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Taobao & Tmall Group of Alibaba","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5159106445","display_name":"Chuan Yu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu, Chuan","raw_affiliation_strings":["Taobao & Tmall Group of Alibaba"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Taobao & Tmall Group of Alibaba","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5159155732","display_name":"Yong Liu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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":0.40230000019073486,"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":0.40230000019073486,"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":0.17599999904632568,"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":0.10670000314712524,"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":0.8669999837875366},{"id":"https://openalex.org/keywords/long-horizon-agents","display_name":"long-horizon agents","score":0.8130000233650208},{"id":"https://openalex.org/keywords/out-of-distribution-generalization","display_name":"out-of-distribution generalization","score":0.9950000047683716},{"id":"https://openalex.org/keywords/attention-drift","display_name":"attention drift","score":0.621999979019165},{"id":"https://openalex.org/keywords/shortcut-learning","display_name":"shortcut learning","score":0.8970000147819519},{"id":"https://openalex.org/keywords/contrastive-regularization","display_name":"contrastive regularization","score":0.8379999995231628}],"concepts":[{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.800599992275238},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7608000040054321},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7109000086784363},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5579000115394592},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.5385000109672546},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4747999906539917},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.4731999933719635},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.4269999861717224},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4146000146865845},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4009000062942505},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.3977000117301941},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.37040001153945923},{"id":"https://openalex.org/C25343380","wikidata":"https://www.wikidata.org/wiki/Q277521","display_name":"Relation (database)","level":2,"score":0.2953000068664551},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.2800000011920929},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.26260000467300415},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.2567000091075897},{"id":"https://openalex.org/C2780665704","wikidata":"https://www.wikidata.org/wiki/Q959298","display_name":"Intervention (counseling)","level":2,"score":0.25290000438690186}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2606.26918","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2606.26918","pdf_url":"https://arxiv.org/pdf/2606.26918","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2606.26918","doi":"https://doi.org/10.48550/arxiv.2606.26918","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.26918","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2606.26918","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2606.26918","pdf_url":"https://arxiv.org/pdf/2606.26918","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"sustainable_development_goals_aurora":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166019693.pdf","grobid_xml":"https://content.openalex.org/works/W7166019693.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W7164897535","https://openalex.org/W7221756309","https://openalex.org/W4296415305","https://openalex.org/W7169888720","https://openalex.org/W7221766770","https://openalex.org/W7214791754","https://openalex.org/W7106213762","https://openalex.org/W7220408055","https://openalex.org/W4415108122","https://openalex.org/W7223018586"],"abstract_inverted_index":{"Language":[0],"agents":[1],"select":[2],"actions":[3],"based":[4],"on":[5,113],"task":[6,16,49,114,141,160],"descriptions,":[7],"environmental":[8],"observations,":[9],"and":[10,34,131,174],"interaction":[11],"histories.":[12],"We":[13],"investigate":[14],"whether":[15,84],"requirements":[17],"reliably":[18],"constrain":[19],"these":[20,79],"decisions":[21],"when":[22,48,59],"the":[23,54,88,151,155,164],"surrounding":[24],"context":[25,166],"already":[26],"suggests":[27],"how":[28],"to":[29,46,61,107,128,136,171],"proceed.":[30],"Across":[31],"five":[32],"models":[33,106,118],"three":[35],"interactive":[36],"environments,":[37],"we":[38,121,143],"find":[39],"that":[40,72,85],"models,":[41],"including":[42],"GPT-5.6-Sol,":[43],"can":[44,100],"continue":[45],"act":[47],"descriptions":[50,161],"are":[51],"replaced":[52],"with":[53,110],"meaningless":[55],"placeholder":[56],"xxxx,":[57],"even":[58],"prompted":[60],"ask":[62],"for":[63],"clarification.":[64],"Action":[65],"generation":[66],"is":[67],"more":[68],"frequent":[69],"in":[70,126,134,178],"states":[71],"suggest":[73,93],"a":[74,94],"concrete":[75],"next":[76],"operation,":[77],"although":[78],"cues":[80,99],"do":[81],"not":[82],"establish":[83],"operation":[86],"serves":[87],"user's":[89],"goal.":[90],"These":[91],"findings":[92],"possible":[95],"training-based":[96],"explanation:":[97],"contextual":[98],"make":[101],"demonstrated":[102],"responses":[103],"predictable,":[104],"allowing":[105],"fit":[108],"them":[109],"limited":[111],"reliance":[112],"requirements.":[115],"Tracking":[116],"seven":[117],"during":[119],"fine-tuning,":[120],"observe":[122],"an":[123,132],"overall":[124],"decrease":[125],"attention":[127,135,170],"task-description":[129,172],"tokens":[130,173],"increase":[133],"current":[137],"observations.":[138],"To":[139],"strengthen":[140],"dependence,":[142],"introduce":[144],"Task-Perturbed":[145],"Negative":[146],"Log-Likelihood":[147],"(TPNLL),":[148],"which":[149],"contrasts":[150],"prediction":[152],"costs":[153],"of":[154],"same":[156],"response":[157],"under":[158],"different":[159],"while":[162],"holding":[163],"non-task":[165],"fixed.":[167],"TPNLL":[168],"increases":[169],"improves":[175],"complete-episode":[176],"success":[177],"out-of-distribution":[179],"settings.":[180]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-06-27T00:00:00"}