{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6ITMBXYHUZOD4T7LEQYKSOX4RN","short_pith_number":"pith:6ITMBXYH","schema_version":"1.0","canonical_sha256":"f226c0df07a65c3e4feb2430a93afc8b7a4930be57954b7cd9ad31ac07697ef1","source":{"kind":"arxiv","id":"2602.01576","version":2},"attestation_state":"computed","paper":{"title":"Generative Visual Code Mobile World Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Jamin Shin, Segyu Lee, Se-Young Yun, Sungjun Han, Woosung Koh","submitted_at":"2026-02-02T03:12:16Z","abstract_excerpt":"Mobile Graphical User Interface (GUI) World Models (WMs) offer a promising path for improving mobile GUI agent performance at train- and inference-time. However, current approaches face a critical trade-off: text-based WMs sacrifice visual fidelity, while the inability of visual WMs in precise text rendering led to their reliance on slow, complex pipelines dependent on numerous external models. We propose a novel paradigm: visual world modeling via renderable code generation, where a single Vision-Language Model (VLM) predicts the next GUI state as executable web code that renders to pixels, r"},"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":"2602.01576","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-02T03:12:16Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"0fab9850127ee8897a48feffc752d524a344fa79f8be4b0c28a9d5d35aee2669","abstract_canon_sha256":"4f378e19b1930465d4745b4adae458005e21a44bb7779457c75261bd304d3502"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:02:31.644410Z","signature_b64":"HYz6rK10Sy/hdEH5VxEmvhIoBwIjxzdaxbT0L7IsGqT4vMl8grc3eaifMZFo2W4cB/B8/pXosuznt/hmM5bFCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f226c0df07a65c3e4feb2430a93afc8b7a4930be57954b7cd9ad31ac07697ef1","last_reissued_at":"2026-05-26T01:02:31.643585Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:02:31.643585Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generative Visual Code Mobile World Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Jamin Shin, Segyu Lee, Se-Young Yun, Sungjun Han, Woosung Koh","submitted_at":"2026-02-02T03:12:16Z","abstract_excerpt":"Mobile Graphical User Interface (GUI) World Models (WMs) offer a promising path for improving mobile GUI agent performance at train- and inference-time. However, current approaches face a critical trade-off: text-based WMs sacrifice visual fidelity, while the inability of visual WMs in precise text rendering led to their reliance on slow, complex pipelines dependent on numerous external models. We propose a novel paradigm: visual world modeling via renderable code generation, where a single Vision-Language Model (VLM) predicts the next GUI state as executable web code that renders to pixels, r"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.01576","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/2602.01576/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":"2602.01576","created_at":"2026-05-26T01:02:31.643699+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.01576v2","created_at":"2026-05-26T01:02:31.643699+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.01576","created_at":"2026-05-26T01:02:31.643699+00:00"},{"alias_kind":"pith_short_12","alias_value":"6ITMBXYHUZOD","created_at":"2026-05-26T01:02:31.643699+00:00"},{"alias_kind":"pith_short_16","alias_value":"6ITMBXYHUZOD4T7L","created_at":"2026-05-26T01:02:31.643699+00:00"},{"alias_kind":"pith_short_8","alias_value":"6ITMBXYH","created_at":"2026-05-26T01:02:31.643699+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":4,"sample":[{"citing_arxiv_id":"2605.10347","citing_title":"How Mobile World Model Guides GUI Agents?","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10347","citing_title":"How Mobile World Model Guides GUI Agents?","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22748","citing_title":"Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond","ref_index":187,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07831","citing_title":"Are GUI Agents Focused Enough? Automated Distraction via Semantic-level UI Element Injection","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6ITMBXYHUZOD4T7LEQYKSOX4RN","json":"https://pith.science/pith/6ITMBXYHUZOD4T7LEQYKSOX4RN.json","graph_json":"https://pith.science/api/pith-number/6ITMBXYHUZOD4T7LEQYKSOX4RN/graph.json","events_json":"https://pith.science/api/pith-number/6ITMBXYHUZOD4T7LEQYKSOX4RN/events.json","paper":"https://pith.science/paper/6ITMBXYH"},"agent_actions":{"view_html":"https://pith.science/pith/6ITMBXYHUZOD4T7LEQYKSOX4RN","download_json":"https://pith.science/pith/6ITMBXYHUZOD4T7LEQYKSOX4RN.json","view_paper":"https://pith.science/paper/6ITMBXYH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.01576&json=true","fetch_graph":"https://pith.science/api/pith-number/6ITMBXYHUZOD4T7LEQYKSOX4RN/graph.json","fetch_events":"https://pith.science/api/pith-number/6ITMBXYHUZOD4T7LEQYKSOX4RN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6ITMBXYHUZOD4T7LEQYKSOX4RN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6ITMBXYHUZOD4T7LEQYKSOX4RN/action/storage_attestation","attest_author":"https://pith.science/pith/6ITMBXYHUZOD4T7LEQYKSOX4RN/action/author_attestation","sign_citation":"https://pith.science/pith/6ITMBXYHUZOD4T7LEQYKSOX4RN/action/citation_signature","submit_replication":"https://pith.science/pith/6ITMBXYHUZOD4T7LEQYKSOX4RN/action/replication_record"}},"created_at":"2026-05-26T01:02:31.643699+00:00","updated_at":"2026-05-26T01:02:31.643699+00:00"}