{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VIRZYC7WV7SVQB7GCYHDSIEE6K","short_pith_number":"pith:VIRZYC7W","schema_version":"1.0","canonical_sha256":"aa239c0bf6afe55807e6160e392084f29de865c069203dfbb6ceb59118172332","source":{"kind":"arxiv","id":"2511.08195","version":4},"attestation_state":"computed","paper":{"title":"UI2Code^N: UI-to-Code Generation as Interactive Visual Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"UI-to-code generation improves by treating it as a closed-loop visual optimization process rather than single-pass output.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiale Cheng, Jie Tang, Mingde Xu, Weihan Wang, Wenyi Hong, Xiaotao Gu, Xinyue Fan, Zhen Yang","submitted_at":"2025-11-11T13:00:09Z","abstract_excerpt":"UI-to-code aims to translate UI screenshots into executable front-end code. Despite progress with vision-language models (VLMs), most existing methods formulate UI-to-code as a single-pass generation, which mismatches real-world UI development that is inherently iterative and feedback-driven. We reformulate UI-to-code as an interactive visual optimization problem, where code generation is embedded in a closed-loop process of execution, visual inspection, and iterative refinement driven by rendered visual feedback. To address the non-differentiability of visual objectives and the noise of absol"},"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":true},"canonical_record":{"source":{"id":"2511.08195","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-11-11T13:00:09Z","cross_cats_sorted":[],"title_canon_sha256":"aff2214936666d9b4032428486b47b37c33b6e0b08c5f2186b21d51894b86bf8","abstract_canon_sha256":"ad02afdc82ccd0c52fa54c78ad4658da6d27cd0928525025a3122ef7173ea72e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:10:29.692054Z","signature_b64":"dQcB4iKnWQ0Ne4X+Pw28jEowUUhevPr5gbRO+xhqkwDM9Cx/lvL+YG8XCny+6fiT3iIkXP/9FPDsMdG5oK3TCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aa239c0bf6afe55807e6160e392084f29de865c069203dfbb6ceb59118172332","last_reissued_at":"2026-06-11T01:10:29.690970Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:10:29.690970Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UI2Code^N: UI-to-Code Generation as Interactive Visual Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"UI-to-code generation improves by treating it as a closed-loop visual optimization process rather than single-pass output.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiale Cheng, Jie Tang, Mingde Xu, Weihan Wang, Wenyi Hong, Xiaotao Gu, Xinyue Fan, Zhen Yang","submitted_at":"2025-11-11T13:00:09Z","abstract_excerpt":"UI-to-code aims to translate UI screenshots into executable front-end code. Despite progress with vision-language models (VLMs), most existing methods formulate UI-to-code as a single-pass generation, which mismatches real-world UI development that is inherently iterative and feedback-driven. We reformulate UI-to-code as an interactive visual optimization problem, where code generation is embedded in a closed-loop process of execution, visual inspection, and iterative refinement driven by rendered visual feedback. To address the non-differentiability of visual objectives and the noise of absol"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experiments demonstrate state-of-the-art performance on UI drafting, UI polishing, and UI editing benchmarks, even outperforming larger models, with performance consistently improving through iterative visual optimization.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That rendered visual feedback can reliably drive iterative refinement despite non-differentiability of visual objectives and noise in absolute evaluators, which RVPO is claimed to address via relative rankings.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"UI2Code^N turns UI-to-code into a closed-loop visual optimization process using Relative Visual Policy Optimization on a 9B model, achieving SOTA results that improve with iterations.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"UI-to-code generation improves by treating it as a closed-loop visual optimization process rather than single-pass output.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"df71cf665818b0354a76b69c10fd51da32223d4911af141200c80cf0cbd33f9c"},"source":{"id":"2511.08195","kind":"arxiv","version":4},"verdict":{"id":"5b955968-47fb-427e-ac63-28ad1ca3fb6e","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-17T23:52:52.575301Z","strongest_claim":"Experiments demonstrate state-of-the-art performance on UI drafting, UI polishing, and UI editing benchmarks, even outperforming larger models, with performance consistently improving through iterative visual optimization.","one_line_summary":"UI2Code^N turns UI-to-code into a closed-loop visual optimization process using Relative Visual Policy Optimization on a 9B model, achieving SOTA results that improve with iterations.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That rendered visual feedback can reliably drive iterative refinement despite non-differentiability of visual objectives and noise in absolute evaluators, which RVPO is claimed to address via relative rankings.","pith_extraction_headline":"UI-to-code generation improves by treating it as a closed-loop visual optimization process rather than single-pass output."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2511.08195/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":2,"snapshot_sha256":"505429aacd6062132bf8add998eaec6b85c0cc0fa3ea5ff43ca59c33a3a87f74"},"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":"2511.08195","created_at":"2026-06-11T01:10:29.691081+00:00"},{"alias_kind":"arxiv_version","alias_value":"2511.08195v4","created_at":"2026-06-11T01:10:29.691081+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.08195","created_at":"2026-06-11T01:10:29.691081+00:00"},{"alias_kind":"pith_short_12","alias_value":"VIRZYC7WV7SV","created_at":"2026-06-11T01:10:29.691081+00:00"},{"alias_kind":"pith_short_16","alias_value":"VIRZYC7WV7SVQB7G","created_at":"2026-06-11T01:10:29.691081+00:00"},{"alias_kind":"pith_short_8","alias_value":"VIRZYC7W","created_at":"2026-06-11T01:10:29.691081+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":3,"sample":[{"citing_arxiv_id":"2606.10334","citing_title":"Self-Distillation Policy Optimization via Visual Feedback: Bridging Code and Visual Artifacts","ref_index":34,"is_internal_anchor":true},{"citing_arxiv_id":"2606.08231","citing_title":"Test-Time Scaling in Multimodal Foundation Models: A Comprehensive Survey of Generation and Reasoning","ref_index":104,"is_internal_anchor":true},{"citing_arxiv_id":"2602.01785","citing_title":"CodeOCR: On the Effectiveness of Vision Language Models in Code Understanding","ref_index":103,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VIRZYC7WV7SVQB7GCYHDSIEE6K","json":"https://pith.science/pith/VIRZYC7WV7SVQB7GCYHDSIEE6K.json","graph_json":"https://pith.science/api/pith-number/VIRZYC7WV7SVQB7GCYHDSIEE6K/graph.json","events_json":"https://pith.science/api/pith-number/VIRZYC7WV7SVQB7GCYHDSIEE6K/events.json","paper":"https://pith.science/paper/VIRZYC7W"},"agent_actions":{"view_html":"https://pith.science/pith/VIRZYC7WV7SVQB7GCYHDSIEE6K","download_json":"https://pith.science/pith/VIRZYC7WV7SVQB7GCYHDSIEE6K.json","view_paper":"https://pith.science/paper/VIRZYC7W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2511.08195&json=true","fetch_graph":"https://pith.science/api/pith-number/VIRZYC7WV7SVQB7GCYHDSIEE6K/graph.json","fetch_events":"https://pith.science/api/pith-number/VIRZYC7WV7SVQB7GCYHDSIEE6K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VIRZYC7WV7SVQB7GCYHDSIEE6K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VIRZYC7WV7SVQB7GCYHDSIEE6K/action/storage_attestation","attest_author":"https://pith.science/pith/VIRZYC7WV7SVQB7GCYHDSIEE6K/action/author_attestation","sign_citation":"https://pith.science/pith/VIRZYC7WV7SVQB7GCYHDSIEE6K/action/citation_signature","submit_replication":"https://pith.science/pith/VIRZYC7WV7SVQB7GCYHDSIEE6K/action/replication_record"}},"created_at":"2026-06-11T01:10:29.691081+00:00","updated_at":"2026-06-11T01:10:29.691081+00:00"}