{"paper":{"title":"AppAgent: Multimodal Agents as Smartphone Users","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"AppAgent lets large language models operate diverse smartphone apps via visual interactions and learns app usage from exploration or demonstrations.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bin Fu, Chi Zhang, Gang Yu, Jiaxuan Liu, Xin Chen, Yucheng Han, Zebiao Huang, Zhao Yang","submitted_at":"2023-12-21T11:52:45Z","abstract_excerpt":"Recent advancements in large language models (LLMs) have led to the creation of intelligent agents capable of performing complex tasks. This paper introduces a novel LLM-based multimodal agent framework designed to operate smartphone applications. Our framework enables the agent to operate smartphone applications through a simplified action space, mimicking human-like interactions such as tapping and swiping. This novel approach bypasses the need for system back-end access, thereby broadening its applicability across diverse apps. Central to our agent's functionality is its innovative learning"},"claims":{"count":3,"items":[{"kind":"strongest_claim","text":"Our framework enables the agent to operate smartphone applications through a simplified action space, mimicking human-like interactions such as tapping and swiping. This novel approach bypasses the need for system back-end access, thereby broadening its applicability across diverse apps.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The agent can reliably learn to navigate and execute tasks in new apps through autonomous exploration or human demonstrations, producing a knowledge base that generalizes across applications.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"AppAgent lets large language models operate diverse smartphone apps via visual interactions and learns app usage from exploration or demonstrations.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"}],"snapshot_sha256":"e0d601a309ea2d10bf0cc1852192cc740bdfac984662b4a09ca1aa64d75b0176"},"source":{"id":"2312.13771","kind":"arxiv","version":2},"verdict":{"id":"8c086a79-6616-4ce5-893d-c4daba15993a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T10:10:48.536796Z","strongest_claim":"Our framework enables the agent to operate smartphone applications through a simplified action space, mimicking human-like interactions such as tapping and swiping. This novel approach bypasses the need for system back-end access, thereby broadening its applicability across diverse apps.","one_line_summary":"AppAgent lets large language models operate diverse smartphone apps via visual interactions and learns app usage from exploration or demonstrations.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The agent can reliably learn to navigate and execute tasks in new apps through autonomous exploration or human demonstrations, producing a knowledge base that generalizes across applications.","pith_extraction_headline":""},"references":{"count":286,"sample":[{"doi":"","year":2022,"title":"Meta FAIR, Anton Bakhtin, Noam Brown, Emily Dinan, Gabriele Farina, Colin Flaherty, Daniel Fried, Andrew Goff, Jonathan Gray, Hengyuan Hu, et al. 2022. Human-level play in the game of diplomacy by com","work_id":"1ceb4a39-65c8-410c-9767-2ac7b120f2e8","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Multimodal web navigation with instruction-finetuned foundation models","work_id":"0f8b8630-9215-4cb8-9b7d-e58e6b1f7bbb","ref_index":6,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"A Real-World WebAgent with Planning, Long Context Understanding, and Program Synthesis","work_id":"0915d1fc-bc46-4128-871e-f9233dca44b6","ref_index":7,"cited_arxiv_id":"2307.12856","is_internal_anchor":true},{"doi":"","year":2023,"title":"Chartllama: A multimodal llm for chart understanding and generation","work_id":"f6c4f1ff-0ce7-47e8-82e5-5c24c2a3ed9f","ref_index":8,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework","work_id":"891b9780-a800-4e3c-bba0-53597ab8dc98","ref_index":9,"cited_arxiv_id":"2308.00352","is_internal_anchor":true}],"resolved_work":286,"snapshot_sha256":"885821595754b102708f214ad9e2814c5947fa18007c701fd0328ce61111a172","internal_anchors":19},"formal_canon":{"evidence_count":2,"snapshot_sha256":"f5de55fee7c9bea80b8fbbb1c74c2316415b5a18f619b190dd3414e0f92b7942"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}