{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZNXTNXANSIVGWVSCS36KJEMBQQ","short_pith_number":"pith:ZNXTNXAN","schema_version":"1.0","canonical_sha256":"cb6f36dc0d922a6b564296fca49181842df86e7abe78b0db0294df208358cafd","source":{"kind":"arxiv","id":"2504.10127","version":2},"attestation_state":"computed","paper":{"title":"Breaking the Data Barrier -- Building GUI Agents Through Task Generalization","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Chang Ma, Junlei Zhang, Junxian He, Qiushi Sun, Zhenzhong Lan, Zichen Ding, Zijie Chen","submitted_at":"2025-04-14T11:35:02Z","abstract_excerpt":"Graphical User Interface (GUI) agents offer cross-platform solutions for automating complex digital tasks, with significant potential to transform productivity workflows. However, their performance is often constrained by the scarcity of high-quality trajectory data. To address this limitation, we propose training Vision Language Models (VLMs) on data-rich, reasoning-intensive tasks during a dedicated mid-training stage, and then examine how incorporating these tasks facilitates generalization to GUI planning scenarios. Specifically, we explore a range of tasks with readily available instructi"},"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":"2504.10127","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-14T11:35:02Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"16e78b223087dc0140198d53e7023d6f055f8b764a2edff14f6c330d21332a00","abstract_canon_sha256":"756192f4a4936e7419703a35effc4f974d4656e1f19a44da69e11b8ac428535f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:49:18.202898Z","signature_b64":"Xd2KuThup/DCbS3bXwKxKmsDpJwOaqUN5DW7G8ITgQ/PCqtxIj8ulFwBs+HmE/T6B0Uxy8D+5r5S8+oX7TJmAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cb6f36dc0d922a6b564296fca49181842df86e7abe78b0db0294df208358cafd","last_reissued_at":"2026-07-05T10:49:18.202404Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:49:18.202404Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Breaking the Data Barrier -- Building GUI Agents Through Task Generalization","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Chang Ma, Junlei Zhang, Junxian He, Qiushi Sun, Zhenzhong Lan, Zichen Ding, Zijie Chen","submitted_at":"2025-04-14T11:35:02Z","abstract_excerpt":"Graphical User Interface (GUI) agents offer cross-platform solutions for automating complex digital tasks, with significant potential to transform productivity workflows. However, their performance is often constrained by the scarcity of high-quality trajectory data. To address this limitation, we propose training Vision Language Models (VLMs) on data-rich, reasoning-intensive tasks during a dedicated mid-training stage, and then examine how incorporating these tasks facilitates generalization to GUI planning scenarios. Specifically, we explore a range of tasks with readily available instructi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.10127","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/2504.10127/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":"2504.10127","created_at":"2026-07-05T10:49:18.202466+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.10127v2","created_at":"2026-07-05T10:49:18.202466+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.10127","created_at":"2026-07-05T10:49:18.202466+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZNXTNXANSIVG","created_at":"2026-07-05T10:49:18.202466+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZNXTNXANSIVGWVSC","created_at":"2026-07-05T10:49:18.202466+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZNXTNXAN","created_at":"2026-07-05T10:49:18.202466+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2508.18265","citing_title":"InternVL3.5: Advancing Open-Source Multimodal Models in Versatility, Reasoning, and Efficiency","ref_index":175,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZNXTNXANSIVGWVSCS36KJEMBQQ","json":"https://pith.science/pith/ZNXTNXANSIVGWVSCS36KJEMBQQ.json","graph_json":"https://pith.science/api/pith-number/ZNXTNXANSIVGWVSCS36KJEMBQQ/graph.json","events_json":"https://pith.science/api/pith-number/ZNXTNXANSIVGWVSCS36KJEMBQQ/events.json","paper":"https://pith.science/paper/ZNXTNXAN"},"agent_actions":{"view_html":"https://pith.science/pith/ZNXTNXANSIVGWVSCS36KJEMBQQ","download_json":"https://pith.science/pith/ZNXTNXANSIVGWVSCS36KJEMBQQ.json","view_paper":"https://pith.science/paper/ZNXTNXAN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.10127&json=true","fetch_graph":"https://pith.science/api/pith-number/ZNXTNXANSIVGWVSCS36KJEMBQQ/graph.json","fetch_events":"https://pith.science/api/pith-number/ZNXTNXANSIVGWVSCS36KJEMBQQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZNXTNXANSIVGWVSCS36KJEMBQQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZNXTNXANSIVGWVSCS36KJEMBQQ/action/storage_attestation","attest_author":"https://pith.science/pith/ZNXTNXANSIVGWVSCS36KJEMBQQ/action/author_attestation","sign_citation":"https://pith.science/pith/ZNXTNXANSIVGWVSCS36KJEMBQQ/action/citation_signature","submit_replication":"https://pith.science/pith/ZNXTNXANSIVGWVSCS36KJEMBQQ/action/replication_record"}},"created_at":"2026-07-05T10:49:18.202466+00:00","updated_at":"2026-07-05T10:49:18.202466+00:00"}