{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HEJ2IV5BR4EDTNXSVGVRWU34WC","short_pith_number":"pith:HEJ2IV5B","schema_version":"1.0","canonical_sha256":"3913a457a18f0839b6f2a9ab1b537cb0bf99f49528c1aeba2d44114880e44fab","source":{"kind":"arxiv","id":"2410.17883","version":2},"attestation_state":"computed","paper":{"title":"Lightweight Neural App Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Filippos Christianos, Georgios Papoudakis, Jianye Hao, Jun Wang, Kun Shao, Thomas Coste","submitted_at":"2024-10-23T13:57:00Z","abstract_excerpt":"This paper introduces a novel mobile phone control architecture, Lightweight Multi-modal App Control (LiMAC), for efficient interactions and control across various Android apps. LiMAC takes as input a textual goal and a sequence of past mobile observations, such as screenshots and corresponding UI trees, to generate precise actions. To address the computational constraints inherent to smartphones, we introduce a small Action Transformer (AcT) integrated with a fine-tuned vision-language model (VLM) for real-time decision-making and task execution. We evaluate LiMAC on two open-source mobile co"},"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":"2410.17883","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-23T13:57:00Z","cross_cats_sorted":[],"title_canon_sha256":"88bc175ff72436eb21bdc0f4c127ba6e68d78dd0a49aa47bd6c7809695cadde2","abstract_canon_sha256":"d7700aeace2e4afe36f4e62b285e2bdb700f0d8b711f2d3ec6eee858da8c9501"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:04.733840Z","signature_b64":"MgqblR48jkm+r9XfDkMv0A1gmAcnmcZE3LarjWNaLTOUFQ9Mti0vhEK1MRAMyyug9gZcBSc3+r4m4+KN8AGgBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3913a457a18f0839b6f2a9ab1b537cb0bf99f49528c1aeba2d44114880e44fab","last_reissued_at":"2026-07-05T10:13:04.733360Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:04.733360Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Lightweight Neural App Control","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Filippos Christianos, Georgios Papoudakis, Jianye Hao, Jun Wang, Kun Shao, Thomas Coste","submitted_at":"2024-10-23T13:57:00Z","abstract_excerpt":"This paper introduces a novel mobile phone control architecture, Lightweight Multi-modal App Control (LiMAC), for efficient interactions and control across various Android apps. LiMAC takes as input a textual goal and a sequence of past mobile observations, such as screenshots and corresponding UI trees, to generate precise actions. To address the computational constraints inherent to smartphones, we introduce a small Action Transformer (AcT) integrated with a fine-tuned vision-language model (VLM) for real-time decision-making and task execution. We evaluate LiMAC on two open-source mobile co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.17883","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/2410.17883/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":"2410.17883","created_at":"2026-07-05T10:13:04.733421+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.17883v2","created_at":"2026-07-05T10:13:04.733421+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.17883","created_at":"2026-07-05T10:13:04.733421+00:00"},{"alias_kind":"pith_short_12","alias_value":"HEJ2IV5BR4ED","created_at":"2026-07-05T10:13:04.733421+00:00"},{"alias_kind":"pith_short_16","alias_value":"HEJ2IV5BR4EDTNXS","created_at":"2026-07-05T10:13:04.733421+00:00"},{"alias_kind":"pith_short_8","alias_value":"HEJ2IV5B","created_at":"2026-07-05T10:13:04.733421+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.17697","citing_title":"Beyond Syntax: Action Semantics Learning for App Agents","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HEJ2IV5BR4EDTNXSVGVRWU34WC","json":"https://pith.science/pith/HEJ2IV5BR4EDTNXSVGVRWU34WC.json","graph_json":"https://pith.science/api/pith-number/HEJ2IV5BR4EDTNXSVGVRWU34WC/graph.json","events_json":"https://pith.science/api/pith-number/HEJ2IV5BR4EDTNXSVGVRWU34WC/events.json","paper":"https://pith.science/paper/HEJ2IV5B"},"agent_actions":{"view_html":"https://pith.science/pith/HEJ2IV5BR4EDTNXSVGVRWU34WC","download_json":"https://pith.science/pith/HEJ2IV5BR4EDTNXSVGVRWU34WC.json","view_paper":"https://pith.science/paper/HEJ2IV5B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.17883&json=true","fetch_graph":"https://pith.science/api/pith-number/HEJ2IV5BR4EDTNXSVGVRWU34WC/graph.json","fetch_events":"https://pith.science/api/pith-number/HEJ2IV5BR4EDTNXSVGVRWU34WC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HEJ2IV5BR4EDTNXSVGVRWU34WC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HEJ2IV5BR4EDTNXSVGVRWU34WC/action/storage_attestation","attest_author":"https://pith.science/pith/HEJ2IV5BR4EDTNXSVGVRWU34WC/action/author_attestation","sign_citation":"https://pith.science/pith/HEJ2IV5BR4EDTNXSVGVRWU34WC/action/citation_signature","submit_replication":"https://pith.science/pith/HEJ2IV5BR4EDTNXSVGVRWU34WC/action/replication_record"}},"created_at":"2026-07-05T10:13:04.733421+00:00","updated_at":"2026-07-05T10:13:04.733421+00:00"}