{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:TBPEPHBCRT4HRNCN4BTFSHLAZF","short_pith_number":"pith:TBPEPHBC","schema_version":"1.0","canonical_sha256":"985e479c228cf878b44de066591d60c94bb495123cd56d6583946da67464df11","source":{"kind":"arxiv","id":"2404.16054","version":2},"attestation_state":"computed","paper":{"title":"LlamaTouch: A Faithful and Scalable Testbed for Mobile UI Task Automation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.HC","authors_text":"Li Zhang, Longxi Gao, Mengwei Xu, Shihe Wang, Xianqing Jia, Yuanchun Li, Yunhe Yan, Zhihan Zheng","submitted_at":"2024-04-12T15:39:09Z","abstract_excerpt":"The emergent large language/multimodal models facilitate the evolution of mobile agents, especially in mobile UI task automation. However, existing evaluation approaches, which rely on human validation or established datasets to compare agent-predicted actions with predefined action sequences, are unscalable and unfaithful. To overcome these limitations, this paper presents LlamaTouch, a testbed for on-device mobile UI task execution and faithful, scalable task evaluation. By observing that the task execution process only transfers UI states, LlamaTouch employs a novel evaluation approach that"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2404.16054","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.HC","submitted_at":"2024-04-12T15:39:09Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"72801d16d0d985fb93d312297ac0452255fe601554d315e78af3662961f99eb1","abstract_canon_sha256":"5eaba51c9902f3be14d8c2206e59e2a774d5df5b559a746a598cebd013899c46"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:51:16.743802Z","signature_b64":"VgMMUU1fIS/vSR5zThFHjzA6nRecHQcYd24kfTd9KzyYRuinEExcjE4Zu99psWKuTlaiE2RGKaOOpmHcEVg/Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"985e479c228cf878b44de066591d60c94bb495123cd56d6583946da67464df11","last_reissued_at":"2026-07-05T08:51:16.743313Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:51:16.743313Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LlamaTouch: A Faithful and Scalable Testbed for Mobile UI Task Automation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.HC","authors_text":"Li Zhang, Longxi Gao, Mengwei Xu, Shihe Wang, Xianqing Jia, Yuanchun Li, Yunhe Yan, Zhihan Zheng","submitted_at":"2024-04-12T15:39:09Z","abstract_excerpt":"The emergent large language/multimodal models facilitate the evolution of mobile agents, especially in mobile UI task automation. However, existing evaluation approaches, which rely on human validation or established datasets to compare agent-predicted actions with predefined action sequences, are unscalable and unfaithful. To overcome these limitations, this paper presents LlamaTouch, a testbed for on-device mobile UI task execution and faithful, scalable task evaluation. By observing that the task execution process only transfers UI states, LlamaTouch employs a novel evaluation approach that"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.16054","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2404.16054/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2404.16054","created_at":"2026-07-05T08:51:16.743389+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.16054v2","created_at":"2026-07-05T08:51:16.743389+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.16054","created_at":"2026-07-05T08:51:16.743389+00:00"},{"alias_kind":"pith_short_12","alias_value":"TBPEPHBCRT4H","created_at":"2026-07-05T08:51:16.743389+00:00"},{"alias_kind":"pith_short_16","alias_value":"TBPEPHBCRT4HRNCN","created_at":"2026-07-05T08:51:16.743389+00:00"},{"alias_kind":"pith_short_8","alias_value":"TBPEPHBC","created_at":"2026-07-05T08:51:16.743389+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2508.16703","citing_title":"ShadowNPU: System and Algorithm Co-design for NPU-Centric On-Device LLM Inference","ref_index":81,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TBPEPHBCRT4HRNCN4BTFSHLAZF","json":"https://pith.science/pith/TBPEPHBCRT4HRNCN4BTFSHLAZF.json","graph_json":"https://pith.science/api/pith-number/TBPEPHBCRT4HRNCN4BTFSHLAZF/graph.json","events_json":"https://pith.science/api/pith-number/TBPEPHBCRT4HRNCN4BTFSHLAZF/events.json","paper":"https://pith.science/paper/TBPEPHBC"},"agent_actions":{"view_html":"https://pith.science/pith/TBPEPHBCRT4HRNCN4BTFSHLAZF","download_json":"https://pith.science/pith/TBPEPHBCRT4HRNCN4BTFSHLAZF.json","view_paper":"https://pith.science/paper/TBPEPHBC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.16054&json=true","fetch_graph":"https://pith.science/api/pith-number/TBPEPHBCRT4HRNCN4BTFSHLAZF/graph.json","fetch_events":"https://pith.science/api/pith-number/TBPEPHBCRT4HRNCN4BTFSHLAZF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TBPEPHBCRT4HRNCN4BTFSHLAZF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TBPEPHBCRT4HRNCN4BTFSHLAZF/action/storage_attestation","attest_author":"https://pith.science/pith/TBPEPHBCRT4HRNCN4BTFSHLAZF/action/author_attestation","sign_citation":"https://pith.science/pith/TBPEPHBCRT4HRNCN4BTFSHLAZF/action/citation_signature","submit_replication":"https://pith.science/pith/TBPEPHBCRT4HRNCN4BTFSHLAZF/action/replication_record"}},"created_at":"2026-07-05T08:51:16.743389+00:00","updated_at":"2026-07-05T08:51:16.743389+00:00"}