{"id":"https://openalex.org/W7155049044","doi":"https://doi.org/10.48550/arxiv.2604.17817","title":"Do LLMs Need to See Everything? A Benchmark and Study of Failures in LLM-driven Smartphone Automation using Screentext vs. Screenshots","display_name":"Do LLMs Need to See Everything? A Benchmark and Study of Failures in LLM-driven Smartphone Automation using Screentext vs. Screenshots","publication_year":2026,"publication_date":"2026-04-20","ids":{"openalex":"https://openalex.org/W7155049044","doi":"https://doi.org/10.48550/arxiv.2604.17817"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.17817","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.17817","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":"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.2604.17817","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134163614","display_name":"Shiquan Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Shiquan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134149909","display_name":"Tianyi Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Tianyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000950850","display_name":"Le Fang","orcid":"https://orcid.org/0000-0003-1860-4008"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang, Le","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134197034","display_name":"Simon D'Alfonso","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"D'Alfonso, Simon","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134147629","display_name":"Hong Jia","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jia, Hong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5015873569","display_name":"Vassilis Kostakos","orcid":"https://orcid.org/0000-0003-2804-6038"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kostakos, Vassilis","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/T12128","display_name":"AI in Service Interactions","score":0.1177000030875206,"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/T12128","display_name":"AI in Service Interactions","score":0.1177000030875206,"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/T12031","display_name":"Speech and dialogue systems","score":0.11010000109672546,"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/T13197","display_name":"Spreadsheets and End-User Computing","score":0.05130000039935112,"subfield":{"id":"https://openalex.org/subfields/1712","display_name":"Software"},"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/benchmark","display_name":"Benchmark (surveying)","score":0.6243000030517578},{"id":"https://openalex.org/keywords/android","display_name":"Android (operating system)","score":0.5965999960899353},{"id":"https://openalex.org/keywords/automation","display_name":"Automation","score":0.48350000381469727},{"id":"https://openalex.org/keywords/phone","display_name":"Phone","score":0.4422999918460846},{"id":"https://openalex.org/keywords/mobile-phone","display_name":"Mobile phone","score":0.41920000314712524},{"id":"https://openalex.org/keywords/mobile-device","display_name":"Mobile device","score":0.37040001153945923},{"id":"https://openalex.org/keywords/mobile-computing","display_name":"Mobile computing","score":0.31119999289512634}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6488999724388123},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6243000030517578},{"id":"https://openalex.org/C557433098","wikidata":"https://www.wikidata.org/wiki/Q94","display_name":"Android (operating system)","level":2,"score":0.5965999960899353},{"id":"https://openalex.org/C115901376","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automation","level":2,"score":0.48350000381469727},{"id":"https://openalex.org/C2778707766","wikidata":"https://www.wikidata.org/wiki/Q202064","display_name":"Phone","level":2,"score":0.4422999918460846},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.4207000136375427},{"id":"https://openalex.org/C2777421447","wikidata":"https://www.wikidata.org/wiki/Q17517","display_name":"Mobile phone","level":2,"score":0.41920000314712524},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.37040001153945923},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.3637000024318695},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.36250001192092896},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.3230000138282776},{"id":"https://openalex.org/C144543869","wikidata":"https://www.wikidata.org/wiki/Q2738570","display_name":"Mobile computing","level":2,"score":0.31119999289512634},{"id":"https://openalex.org/C2776543023","wikidata":"https://www.wikidata.org/wiki/Q2147046","display_name":"Rework","level":2,"score":0.3010999858379364},{"id":"https://openalex.org/C3020250448","wikidata":"https://www.wikidata.org/wiki/Q620615","display_name":"Smartphone application","level":2,"score":0.29989999532699585},{"id":"https://openalex.org/C2779018934","wikidata":"https://www.wikidata.org/wiki/Q1129653","display_name":"Everyday life","level":2,"score":0.2996000051498413},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2953999936580658},{"id":"https://openalex.org/C62230096","wikidata":"https://www.wikidata.org/wiki/Q275969","display_name":"Crowdsourcing","level":2,"score":0.2842000126838684},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.26649999618530273},{"id":"https://openalex.org/C187191949","wikidata":"https://www.wikidata.org/wiki/Q1138496","display_name":"Profiling (computer programming)","level":2,"score":0.2590000033378601},{"id":"https://openalex.org/C2988145974","wikidata":"https://www.wikidata.org/wiki/Q620615","display_name":"Mobile apps","level":2,"score":0.2535000145435333}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.17817","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.17817","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":"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.2604.17817","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.17817","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":"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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"With":[0],"the":[1],"rapid":[2],"advancement":[3],"of":[4,37,63,119],"large":[5],"language":[6],"models":[7],"(LLMs),":[8],"mobile":[9,139],"agents":[10,30],"have":[11],"emerged":[12],"as":[13],"promising":[14],"tools":[15],"for":[16,137],"phone":[17],"automation,":[18],"simulating":[19],"human":[20],"interactions":[21],"on":[22,42,93],"screens":[23],"to":[24,77],"accomplish":[25],"complex":[26],"tasks.":[27],"However,":[28],"these":[29],"often":[31],"suffer":[32],"from":[33],"low":[34],"accuracy,":[35],"misinterpretation":[36],"user":[38],"instructions,":[39],"and":[40,51,87,95,132,142],"failure":[41,113],"challenging":[43],"tasks,":[44],"with":[45,103],"limited":[46],"prior":[47],"work":[48],"examining":[49],"why":[50],"where":[52],"they":[53],"fail.":[54],"To":[55],"address":[56],"this,":[57],"we":[58,115],"introduce":[59],"DailyDroid,":[60],"a":[61,117],"benchmark":[62],"75":[64],"tasks":[65],"in":[66,127],"five":[67],"scenarios":[68],"across":[69,97],"25":[70],"Android":[71],"apps,":[72],"spanning":[73],"three":[74],"difficulty":[75],"levels":[76],"mimic":[78],"everyday":[79],"smartphone":[80],"use.":[81],"We":[82],"evaluate":[83],"it":[84],"using":[85],"text-only":[86],"multimodal":[88,104],"(text":[89],"+":[90],"screenshot)":[91],"inputs":[92,105],"GPT-4o":[94],"o4-mini":[96],"300":[98],"trials,":[99],"revealing":[100],"comparable":[101],"performance":[102],"yielding":[106],"marginally":[107],"higher":[108],"success":[109],"rates.":[110],"Through":[111],"in-depth":[112],"analysis,":[114],"compile":[116],"handbook":[118],"common":[120],"failures.":[121],"Our":[122],"findings":[123],"reveal":[124],"critical":[125],"issues":[126],"UI":[128,143],"accessibility,":[129],"input":[130],"modalities,":[131],"LLM/app":[133],"design,":[134],"offering":[135],"implications":[136],"future":[138],"agents,":[140],"applications,":[141],"development.":[144]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-22T00:00:00"}