{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:5AOAF5FJ7EKD6XHF65HDILJ3NL","short_pith_number":"pith:5AOAF5FJ","schema_version":"1.0","canonical_sha256":"e81c02f4a9f9143f5ce5f74e342d3b6af1b808fa89e64f38c286d9d29a9438ba","source":{"kind":"arxiv","id":"2402.15307","version":1},"attestation_state":"computed","paper":{"title":"Representing Online Handwriting for Recognition in Large Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Anastasiia Fadeeva, Andrii Maksai, Claudiu Musat, Efi Kokiopoulou, Jesse Berent, Mark Collier, Philippe Schlattner","submitted_at":"2024-02-23T13:11:10Z","abstract_excerpt":"The adoption of tablets with touchscreens and styluses is increasing, and a key feature is converting handwriting to text, enabling search, indexing, and AI assistance. Meanwhile, vision-language models (VLMs) are now the go-to solution for image understanding, thanks to both their state-of-the-art performance across a variety of tasks and the simplicity of a unified approach to training, fine-tuning, and inference. While VLMs obtain high performance on image-based tasks, they perform poorly on handwriting recognition when applied naively, i.e., by rendering handwriting as an image and perform"},"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":"2402.15307","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-02-23T13:11:10Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"791042c7c549081640945ffd4e837d5fb9708d154b9191a7ef206e919700316a","abstract_canon_sha256":"ceffd588c16f1154f0f4a55f656a9fbb7cf098fc35ad1ed96d3f63d3862267e5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:48:39.059003Z","signature_b64":"Nu3dUhxkAF89oK5Da6b/6EPR/H5Z1SN36mlx4g5cNlUcOSY9mHp66H4AkNtoSz4EMdFvSHp7PrwMDJk6aDLuDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e81c02f4a9f9143f5ce5f74e342d3b6af1b808fa89e64f38c286d9d29a9438ba","last_reissued_at":"2026-07-05T07:48:39.058568Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:48:39.058568Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Representing Online Handwriting for Recognition in Large Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Anastasiia Fadeeva, Andrii Maksai, Claudiu Musat, Efi Kokiopoulou, Jesse Berent, Mark Collier, Philippe Schlattner","submitted_at":"2024-02-23T13:11:10Z","abstract_excerpt":"The adoption of tablets with touchscreens and styluses is increasing, and a key feature is converting handwriting to text, enabling search, indexing, and AI assistance. Meanwhile, vision-language models (VLMs) are now the go-to solution for image understanding, thanks to both their state-of-the-art performance across a variety of tasks and the simplicity of a unified approach to training, fine-tuning, and inference. While VLMs obtain high performance on image-based tasks, they perform poorly on handwriting recognition when applied naively, i.e., by rendering handwriting as an image and perform"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.15307","kind":"arxiv","version":1},"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/2402.15307/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":"2402.15307","created_at":"2026-07-05T07:48:39.058627+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.15307v1","created_at":"2026-07-05T07:48:39.058627+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.15307","created_at":"2026-07-05T07:48:39.058627+00:00"},{"alias_kind":"pith_short_12","alias_value":"5AOAF5FJ7EKD","created_at":"2026-07-05T07:48:39.058627+00:00"},{"alias_kind":"pith_short_16","alias_value":"5AOAF5FJ7EKD6XHF","created_at":"2026-07-05T07:48:39.058627+00:00"},{"alias_kind":"pith_short_8","alias_value":"5AOAF5FJ","created_at":"2026-07-05T07:48:39.058627+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.11477","citing_title":"Towards Fully Automated Exam Grading: Fairness-Aware Recognition of Handwritten Answers with Foundation Models","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5AOAF5FJ7EKD6XHF65HDILJ3NL","json":"https://pith.science/pith/5AOAF5FJ7EKD6XHF65HDILJ3NL.json","graph_json":"https://pith.science/api/pith-number/5AOAF5FJ7EKD6XHF65HDILJ3NL/graph.json","events_json":"https://pith.science/api/pith-number/5AOAF5FJ7EKD6XHF65HDILJ3NL/events.json","paper":"https://pith.science/paper/5AOAF5FJ"},"agent_actions":{"view_html":"https://pith.science/pith/5AOAF5FJ7EKD6XHF65HDILJ3NL","download_json":"https://pith.science/pith/5AOAF5FJ7EKD6XHF65HDILJ3NL.json","view_paper":"https://pith.science/paper/5AOAF5FJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.15307&json=true","fetch_graph":"https://pith.science/api/pith-number/5AOAF5FJ7EKD6XHF65HDILJ3NL/graph.json","fetch_events":"https://pith.science/api/pith-number/5AOAF5FJ7EKD6XHF65HDILJ3NL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5AOAF5FJ7EKD6XHF65HDILJ3NL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5AOAF5FJ7EKD6XHF65HDILJ3NL/action/storage_attestation","attest_author":"https://pith.science/pith/5AOAF5FJ7EKD6XHF65HDILJ3NL/action/author_attestation","sign_citation":"https://pith.science/pith/5AOAF5FJ7EKD6XHF65HDILJ3NL/action/citation_signature","submit_replication":"https://pith.science/pith/5AOAF5FJ7EKD6XHF65HDILJ3NL/action/replication_record"}},"created_at":"2026-07-05T07:48:39.058627+00:00","updated_at":"2026-07-05T07:48:39.058627+00:00"}