{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FJLJZS66WZSA7RVUEARY26O2IS","short_pith_number":"pith:FJLJZS66","schema_version":"1.0","canonical_sha256":"2a569ccbdeb6640fc6b420238d79da44aa5a7afc0435d40fd5addc0ed1faa205","source":{"kind":"arxiv","id":"2406.03679","version":6},"attestation_state":"computed","paper":{"title":"On the Effects of Data Scale on UI Control Agents","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Alice Li, Chris Rawles, Divya Tyamagundlu, Folawiyo Campbell-Ajala, Oriana Riva, Wei Li, William Bishop","submitted_at":"2024-06-06T01:49:29Z","abstract_excerpt":"Autonomous agents that control computer interfaces to accomplish human tasks are emerging. Leveraging LLMs to power such agents has been of special interest, but unless fine-tuned on human-collected task demonstrations, performance is still relatively low. In this work we study whether fine-tuning alone is a viable approach for building real-world computer control agents. In particularly, we investigate how performance measured on both high and low-level tasks in domain and out of domain scales as more training data is collected. To this end we collect and release a new dataset, AndroidControl"},"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":"2406.03679","kind":"arxiv","version":6},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.AI","submitted_at":"2024-06-06T01:49:29Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9d86201466aec93d39c420e3140bad7c6c30a02c986ff7fb9f0772ee854d391a","abstract_canon_sha256":"e33a50daffdca9c2cf528d8ff53cc5ae7069ae552ebc98bf967198e4b3908a9f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:34:56.928064Z","signature_b64":"tLjjoHxObORyyE8Rc4pAcu8oAIik45M4YmmFWWih88QTD8Olonz6JfitcpOFYQti3FoBvN3s8xZ77sWyVv0iCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2a569ccbdeb6640fc6b420238d79da44aa5a7afc0435d40fd5addc0ed1faa205","last_reissued_at":"2026-07-05T09:34:56.927592Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:34:56.927592Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On the Effects of Data Scale on UI Control Agents","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Alice Li, Chris Rawles, Divya Tyamagundlu, Folawiyo Campbell-Ajala, Oriana Riva, Wei Li, William Bishop","submitted_at":"2024-06-06T01:49:29Z","abstract_excerpt":"Autonomous agents that control computer interfaces to accomplish human tasks are emerging. Leveraging LLMs to power such agents has been of special interest, but unless fine-tuned on human-collected task demonstrations, performance is still relatively low. In this work we study whether fine-tuning alone is a viable approach for building real-world computer control agents. In particularly, we investigate how performance measured on both high and low-level tasks in domain and out of domain scales as more training data is collected. To this end we collect and release a new dataset, AndroidControl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.03679","kind":"arxiv","version":6},"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/2406.03679/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":"2406.03679","created_at":"2026-07-05T09:34:56.927648+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.03679v6","created_at":"2026-07-05T09:34:56.927648+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.03679","created_at":"2026-07-05T09:34:56.927648+00:00"},{"alias_kind":"pith_short_12","alias_value":"FJLJZS66WZSA","created_at":"2026-07-05T09:34:56.927648+00:00"},{"alias_kind":"pith_short_16","alias_value":"FJLJZS66WZSA7RVU","created_at":"2026-07-05T09:34:56.927648+00:00"},{"alias_kind":"pith_short_8","alias_value":"FJLJZS66","created_at":"2026-07-05T09:34:56.927648+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":19,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.31270","citing_title":"Learning from Failure: Inference-Time Self-Improvement for Computer-Use Agents","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07358","citing_title":"A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applications","ref_index":121,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25160","citing_title":"ScaleWoB: Guiding GUI Agents with Coding Agents via Large-Scale Environmental Synthesis","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30084","citing_title":"One Forward Beats Two: InnerZoom for Accurate and Efficient GUI Grounding","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29705","citing_title":"GUICrafter: Weakly-Supervised GUI Agent Leveraging Massive Unannotated Screenshots","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28629","citing_title":"Mobile-Aptus: Confidence-Driven Proactive and Robust Interaction in MLLM-based Mobile-Using Agents","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2501.16150","citing_title":"A Comprehensive Survey of Agents for Computer Use: Foundations, Challenges, and Future Directions","ref_index":87,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07358","citing_title":"A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applications","ref_index":129,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16883","citing_title":"SE-GA: Memory-Augmented Self-Evolution for GUI Agents","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2506.17697","citing_title":"Beyond Syntax: Action Semantics Learning for App Agents","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2412.04454","citing_title":"Aguvis: Unified Pure Vision Agents for Autonomous GUI Interaction","ref_index":91,"is_internal_anchor":false},{"citing_arxiv_id":"2504.10458","citing_title":"GUI-R1 : A Generalist R1-Style Vision-Language Action Model For GUI Agents","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2409.17146","citing_title":"Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2409.07429","citing_title":"Agent Workflow Memory","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11212","citing_title":"ReVision: Scaling Computer-Use Agents via Temporal Visual Redundancy Reduction","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2405.14573","citing_title":"AndroidWorld: A Dynamic Benchmarking Environment for Autonomous Agents","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2410.23218","citing_title":"OS-ATLAS: A Foundation Action Model for Generalist GUI Agents","ref_index":114,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11212","citing_title":"ReVision: Scaling Computer-Use Agents via Temporal Visual Redundancy Reduction","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07358","citing_title":"A Comprehensive Survey on Agent Skills: Taxonomy, Techniques, and Applications","ref_index":127,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FJLJZS66WZSA7RVUEARY26O2IS","json":"https://pith.science/pith/FJLJZS66WZSA7RVUEARY26O2IS.json","graph_json":"https://pith.science/api/pith-number/FJLJZS66WZSA7RVUEARY26O2IS/graph.json","events_json":"https://pith.science/api/pith-number/FJLJZS66WZSA7RVUEARY26O2IS/events.json","paper":"https://pith.science/paper/FJLJZS66"},"agent_actions":{"view_html":"https://pith.science/pith/FJLJZS66WZSA7RVUEARY26O2IS","download_json":"https://pith.science/pith/FJLJZS66WZSA7RVUEARY26O2IS.json","view_paper":"https://pith.science/paper/FJLJZS66","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.03679&json=true","fetch_graph":"https://pith.science/api/pith-number/FJLJZS66WZSA7RVUEARY26O2IS/graph.json","fetch_events":"https://pith.science/api/pith-number/FJLJZS66WZSA7RVUEARY26O2IS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FJLJZS66WZSA7RVUEARY26O2IS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FJLJZS66WZSA7RVUEARY26O2IS/action/storage_attestation","attest_author":"https://pith.science/pith/FJLJZS66WZSA7RVUEARY26O2IS/action/author_attestation","sign_citation":"https://pith.science/pith/FJLJZS66WZSA7RVUEARY26O2IS/action/citation_signature","submit_replication":"https://pith.science/pith/FJLJZS66WZSA7RVUEARY26O2IS/action/replication_record"}},"created_at":"2026-07-05T09:34:56.927648+00:00","updated_at":"2026-07-05T09:34:56.927648+00:00"}