{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:PADQJG66RQ73J2J5GHB5OIIE7W","short_pith_number":"pith:PADQJG66","canonical_record":{"source":{"id":"2605.14708","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:24:44Z","cross_cats_sorted":[],"title_canon_sha256":"7e9d295f264f56563720097f63d96edad649716d38bfc6814486f3553f853f77","abstract_canon_sha256":"0667db19a66b31850f0dd9d8cdeb6d60eb029b8f0216dd0ab767a44e95764e7a"},"schema_version":"1.0"},"canonical_sha256":"7807049bde8c3fb4e93d31c3d72104fdbb4253a1b27fec316ee732cc27f285e8","source":{"kind":"arxiv","id":"2605.14708","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14708","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14708v1","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14708","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"pith_short_12","alias_value":"PADQJG66RQ73","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"PADQJG66RQ73J2J5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"PADQJG66","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:PADQJG66RQ73J2J5GHB5OIIE7W","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14708","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:24:44Z","cross_cats_sorted":[],"title_canon_sha256":"7e9d295f264f56563720097f63d96edad649716d38bfc6814486f3553f853f77","abstract_canon_sha256":"0667db19a66b31850f0dd9d8cdeb6d60eb029b8f0216dd0ab767a44e95764e7a"},"schema_version":"1.0"},"canonical_sha256":"7807049bde8c3fb4e93d31c3d72104fdbb4253a1b27fec316ee732cc27f285e8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:59.246572Z","signature_b64":"CrR8tTvH+Gxo6q2hESQprJzXNDlMgMx5UvmKPJ4BeRiEjhbJfIgEZPQfZ4ERwYr0V+l+48hQBc3Vl+FeBg1ADQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7807049bde8c3fb4e93d31c3d72104fdbb4253a1b27fec316ee732cc27f285e8","last_reissued_at":"2026-05-17T23:38:59.246059Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:59.246059Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14708","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tOfym4sFQwLeLNJLV8IM1l1lsPTbhiA7FATMbN3gf5FFc68KEfWDiMwX92VBy5cC25zCFwvd9VqzBquMr2kMCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T14:38:33.753209Z"},"content_sha256":"585382907495e57aa3eedab1edbb98ed7b083a86c0d6083af330f0307fda3fba","schema_version":"1.0","event_id":"sha256:585382907495e57aa3eedab1edbb98ed7b083a86c0d6083af330f0307fda3fba"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:PADQJG66RQ73J2J5GHB5OIIE7W","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"StyleTextGen: Style-Conditioned Multilingual Scene Text Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"StyleTextGen generates scene text that matches reference visual styles across languages using a dedicated dual-branch encoder.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fangmin Zhao, Liu Yu, Yan Shu, Yichao Liu, Yu Zhou, Zeyu Chen","submitted_at":"2026-05-14T11:24:44Z","abstract_excerpt":"Style-conditioned scene text generation faces unique challenges in extracting precise text styles from complex backgrounds and maintaining fine-grained style consistency across characters, especially for multilingual scripts. We propose StyleTextGen, a novel framework that learns to perceive and replicate visual text styles across different languages and writing systems. Our approach features three key contributions: First, we introduce a dual-branch style encoder dedicated to style modeling, yielding robust multilingual text style representations in complex real-world scenes. Second, we desig"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"StyleTextGen significantly outperforms existing methods in style consistency and cross-lingual generalization, establishing new state-of-the-art performance in multilingual style-conditioned text generation.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The dual-branch style encoder and consistency loss successfully extract and maintain precise, fine-grained text styles from complex real-world backgrounds across languages without post-hoc tuning or dataset-specific adjustments.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"StyleTextGen proposes a dual-branch style encoder, text style consistency loss, and mask-guided inference to achieve superior style consistency and cross-lingual performance in multilingual scene text generation on a new bilingual benchmark.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"StyleTextGen generates scene text that matches reference visual styles across languages using a dedicated dual-branch encoder.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a4a58b41c0177bef8849557391938fcef7650b084975e924aead69aeac53d800"},"source":{"id":"2605.14708","kind":"arxiv","version":1},"verdict":{"id":"86bc7873-b64a-4568-9f04-356d2988cb97","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T04:52:00.033596Z","strongest_claim":"StyleTextGen significantly outperforms existing methods in style consistency and cross-lingual generalization, establishing new state-of-the-art performance in multilingual style-conditioned text generation.","one_line_summary":"StyleTextGen proposes a dual-branch style encoder, text style consistency loss, and mask-guided inference to achieve superior style consistency and cross-lingual performance in multilingual scene text generation on a new bilingual benchmark.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The dual-branch style encoder and consistency loss successfully extract and maintain precise, fine-grained text styles from complex real-world backgrounds across languages without post-hoc tuning or dataset-specific adjustments.","pith_extraction_headline":"StyleTextGen generates scene text that matches reference visual styles across languages using a dedicated dual-branch encoder."},"references":{"count":61,"sample":[{"doi":"","year":2023,"title":"In- structpix2pix: Learning to follow image editing instructions","work_id":"273977e6-dd4b-4c1d-9f32-f13a08c04d96","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"The devil is in fine-tuning and long-tailed prob- lems: a new benchmark for scene text detection.arXiv preprint arXiv:2505.15649, 2025","work_id":"f77ba1f6-e8b7-4bf2-b13d-82139f5b7dbc","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Posta: A go-to framework for customized artistic poster gen- eration","work_id":"8bdfaaf7-a8a9-4f09-b68a-bebb86151d83","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Textdiffuser-2: Unleashing the power of language models for text rendering.arXiv preprint arXiv:2311.16465, 2023","work_id":"8cc07be4-b59c-466e-b152-0a92bc8ca0d6","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"arXiv preprint arXiv:2305.10855 (2023) 7","work_id":"f43372f1-5688-4262-a5f2-2e8f7da8d54c","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":61,"snapshot_sha256":"8cafd2d22f8cc8ad8e011b1972cac2d363c628095747ff05e4025ce4f2f76597","internal_anchors":3},"formal_canon":{"evidence_count":2,"snapshot_sha256":"189af8e91dcf8cfd3df5ab944e9692e818db19b4ba5f23ace4f8e6964d00ed14"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"86bc7873-b64a-4568-9f04-356d2988cb97"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:38:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"n9PvDd68OaiZNJv8Dd89VX+hvTg5iidMtjk0/L4EF0eeUToHcDNFrJrAZxclsNr+Pmpg/DcHs/foXAf/CuD8Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T14:38:33.754456Z"},"content_sha256":"f9cb72d6b9d38ca77dd1b0dc816ec427b454b78a1315ea4daa2de0dfad15dbbb","schema_version":"1.0","event_id":"sha256:f9cb72d6b9d38ca77dd1b0dc816ec427b454b78a1315ea4daa2de0dfad15dbbb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PADQJG66RQ73J2J5GHB5OIIE7W/bundle.json","state_url":"https://pith.science/pith/PADQJG66RQ73J2J5GHB5OIIE7W/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PADQJG66RQ73J2J5GHB5OIIE7W/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-30T14:38:33Z","links":{"resolver":"https://pith.science/pith/PADQJG66RQ73J2J5GHB5OIIE7W","bundle":"https://pith.science/pith/PADQJG66RQ73J2J5GHB5OIIE7W/bundle.json","state":"https://pith.science/pith/PADQJG66RQ73J2J5GHB5OIIE7W/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PADQJG66RQ73J2J5GHB5OIIE7W/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:PADQJG66RQ73J2J5GHB5OIIE7W","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"0667db19a66b31850f0dd9d8cdeb6d60eb029b8f0216dd0ab767a44e95764e7a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:24:44Z","title_canon_sha256":"7e9d295f264f56563720097f63d96edad649716d38bfc6814486f3553f853f77"},"schema_version":"1.0","source":{"id":"2605.14708","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14708","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14708v1","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14708","created_at":"2026-05-17T23:38:59Z"},{"alias_kind":"pith_short_12","alias_value":"PADQJG66RQ73","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"PADQJG66RQ73J2J5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"PADQJG66","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:f9cb72d6b9d38ca77dd1b0dc816ec427b454b78a1315ea4daa2de0dfad15dbbb","target":"graph","created_at":"2026-05-17T23:38:59Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"StyleTextGen significantly outperforms existing methods in style consistency and cross-lingual generalization, establishing new state-of-the-art performance in multilingual style-conditioned text generation."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The dual-branch style encoder and consistency loss successfully extract and maintain precise, fine-grained text styles from complex real-world backgrounds across languages without post-hoc tuning or dataset-specific adjustments."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"StyleTextGen proposes a dual-branch style encoder, text style consistency loss, and mask-guided inference to achieve superior style consistency and cross-lingual performance in multilingual scene text generation on a new bilingual benchmark."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"StyleTextGen generates scene text that matches reference visual styles across languages using a dedicated dual-branch encoder."}],"snapshot_sha256":"a4a58b41c0177bef8849557391938fcef7650b084975e924aead69aeac53d800"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"189af8e91dcf8cfd3df5ab944e9692e818db19b4ba5f23ace4f8e6964d00ed14"},"paper":{"abstract_excerpt":"Style-conditioned scene text generation faces unique challenges in extracting precise text styles from complex backgrounds and maintaining fine-grained style consistency across characters, especially for multilingual scripts. We propose StyleTextGen, a novel framework that learns to perceive and replicate visual text styles across different languages and writing systems. Our approach features three key contributions: First, we introduce a dual-branch style encoder dedicated to style modeling, yielding robust multilingual text style representations in complex real-world scenes. Second, we desig","authors_text":"Fangmin Zhao, Liu Yu, Yan Shu, Yichao Liu, Yu Zhou, Zeyu Chen","cross_cats":[],"headline":"StyleTextGen generates scene text that matches reference visual styles across languages using a dedicated dual-branch encoder.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:24:44Z","title":"StyleTextGen: Style-Conditioned Multilingual Scene Text Generation"},"references":{"count":61,"internal_anchors":3,"resolved_work":61,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"In- structpix2pix: Learning to follow image editing instructions","work_id":"273977e6-dd4b-4c1d-9f32-f13a08c04d96","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"The devil is in fine-tuning and long-tailed prob- lems: a new benchmark for scene text detection.arXiv preprint arXiv:2505.15649, 2025","work_id":"f77ba1f6-e8b7-4bf2-b13d-82139f5b7dbc","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Posta: A go-to framework for customized artistic poster gen- eration","work_id":"8bdfaaf7-a8a9-4f09-b68a-bebb86151d83","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Textdiffuser-2: Unleashing the power of language models for text rendering.arXiv preprint arXiv:2311.16465, 2023","work_id":"8cc07be4-b59c-466e-b152-0a92bc8ca0d6","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"arXiv preprint arXiv:2305.10855 (2023) 7","work_id":"f43372f1-5688-4262-a5f2-2e8f7da8d54c","year":2023}],"snapshot_sha256":"8cafd2d22f8cc8ad8e011b1972cac2d363c628095747ff05e4025ce4f2f76597"},"source":{"id":"2605.14708","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T04:52:00.033596Z","id":"86bc7873-b64a-4568-9f04-356d2988cb97","model_set":{"reader":"grok-4.3"},"one_line_summary":"StyleTextGen proposes a dual-branch style encoder, text style consistency loss, and mask-guided inference to achieve superior style consistency and cross-lingual performance in multilingual scene text generation on a new bilingual benchmark.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"StyleTextGen generates scene text that matches reference visual styles across languages using a dedicated dual-branch encoder.","strongest_claim":"StyleTextGen significantly outperforms existing methods in style consistency and cross-lingual generalization, establishing new state-of-the-art performance in multilingual style-conditioned text generation.","weakest_assumption":"The dual-branch style encoder and consistency loss successfully extract and maintain precise, fine-grained text styles from complex real-world backgrounds across languages without post-hoc tuning or dataset-specific adjustments."}},"verdict_id":"86bc7873-b64a-4568-9f04-356d2988cb97"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:585382907495e57aa3eedab1edbb98ed7b083a86c0d6083af330f0307fda3fba","target":"record","created_at":"2026-05-17T23:38:59Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"0667db19a66b31850f0dd9d8cdeb6d60eb029b8f0216dd0ab767a44e95764e7a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-14T11:24:44Z","title_canon_sha256":"7e9d295f264f56563720097f63d96edad649716d38bfc6814486f3553f853f77"},"schema_version":"1.0","source":{"id":"2605.14708","kind":"arxiv","version":1}},"canonical_sha256":"7807049bde8c3fb4e93d31c3d72104fdbb4253a1b27fec316ee732cc27f285e8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7807049bde8c3fb4e93d31c3d72104fdbb4253a1b27fec316ee732cc27f285e8","first_computed_at":"2026-05-17T23:38:59.246059Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:59.246059Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CrR8tTvH+Gxo6q2hESQprJzXNDlMgMx5UvmKPJ4BeRiEjhbJfIgEZPQfZ4ERwYr0V+l+48hQBc3Vl+FeBg1ADQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:59.246572Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14708","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:585382907495e57aa3eedab1edbb98ed7b083a86c0d6083af330f0307fda3fba","sha256:f9cb72d6b9d38ca77dd1b0dc816ec427b454b78a1315ea4daa2de0dfad15dbbb"],"state_sha256":"14793c68e3e8a310aac933804fa889d22a9e29019c393b7df7520542926f8f5e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iRX5iM9qJFQ3BiUWC2u6vDFOlylvLvfnOq1bt6olcPNrh1COXNH52xxbF6t5UaYtAUepnTTKG0XPO/tsmrapBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T14:38:33.759852Z","bundle_sha256":"af3c4b89c7c0372851c6572049eb4b11da3351fe29493fb14c428cdb1ca2db97"}}