{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:BVB34ZNP3LM3KVFFYMO6GPIYRV","short_pith_number":"pith:BVB34ZNP","canonical_record":{"source":{"id":"2503.15195","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-19T13:33:29Z","cross_cats_sorted":[],"title_canon_sha256":"7df2764720634f52c55509dc9f51921a4eabec884184655d80de5e10dafb322f","abstract_canon_sha256":"a310394a3d57bf0021f71b9d2a87042aa64a8eaa2080999776dd84e53d88d585"},"schema_version":"1.0"},"canonical_sha256":"0d43be65afdad9b554a5c31de33d188d75515c38caa7344fd7ed0f0d7474c783","source":{"kind":"arxiv","id":"2503.15195","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.15195","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"arxiv_version","alias_value":"2503.15195v3","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.15195","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_12","alias_value":"BVB34ZNP3LM3","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_16","alias_value":"BVB34ZNP3LM3KVFF","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_8","alias_value":"BVB34ZNP","created_at":"2026-07-05T11:25:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:BVB34ZNP3LM3KVFFYMO6GPIYRV","target":"record","payload":{"canonical_record":{"source":{"id":"2503.15195","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-19T13:33:29Z","cross_cats_sorted":[],"title_canon_sha256":"7df2764720634f52c55509dc9f51921a4eabec884184655d80de5e10dafb322f","abstract_canon_sha256":"a310394a3d57bf0021f71b9d2a87042aa64a8eaa2080999776dd84e53d88d585"},"schema_version":"1.0"},"canonical_sha256":"0d43be65afdad9b554a5c31de33d188d75515c38caa7344fd7ed0f0d7474c783","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:42.914648Z","signature_b64":"2UvMsSjSODvunFB1MuH90pS3IR25rj85JqmGu3hZcg7dsa/W28hqn7glB3BWJ806oH7+i0bVVT4EcUEvkfbqAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0d43be65afdad9b554a5c31de33d188d75515c38caa7344fd7ed0f0d7474c783","last_reissued_at":"2026-07-05T11:25:42.914213Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:42.914213Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.15195","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:25:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0MystevLqML4+AYGt8fOonS31AVTfbixtw4Kqiq+xOrPuzk/WcyjFqqH4boYVAUczeP+QXFgGNmfJznOcxP0AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T20:12:05.908621Z"},"content_sha256":"cf2b9d0b3bbb3860fa15bd78853841e35763cfa6ed746eb96f15bc572ed13482","schema_version":"1.0","event_id":"sha256:cf2b9d0b3bbb3860fa15bd78853841e35763cfa6ed746eb96f15bc572ed13482"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:BVB34ZNP3LM3KVFFYMO6GPIYRV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Benchmarking Large Language Models for Handwritten Text Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Giorgia Crosilla, Giovanni Colavizza, Lukas Klic","submitted_at":"2025-03-19T13:33:29Z","abstract_excerpt":"Traditional machine learning models for Handwritten Text Recognition (HTR) rely on supervised training, requiring extensive manual annotations, and often produce errors due to the separation between layout and text processing. In contrast, Multimodal Large Language Models (MLLMs) offer a general approach to recognizing diverse handwriting styles without the need for model-specific training. The study benchmarks various proprietary and open-source LLMs against Transkribus models, evaluating their performance on both modern and historical datasets written in English, French, German, and Italian."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.15195","kind":"arxiv","version":3},"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/2503.15195/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:25:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q2aYCG3AbSa/FDP+fB6UBM4jGGPHvyyYBmfjeZzTv6Md7fSgTrBtWuNNIAuA+cmZCXfSYmGhwitiTjI3OYf5Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T20:12:05.909251Z"},"content_sha256":"8cceff09ca22879e876c377e6f2656e849aaac220819748ff805987810277502","schema_version":"1.0","event_id":"sha256:8cceff09ca22879e876c377e6f2656e849aaac220819748ff805987810277502"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BVB34ZNP3LM3KVFFYMO6GPIYRV/bundle.json","state_url":"https://pith.science/pith/BVB34ZNP3LM3KVFFYMO6GPIYRV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BVB34ZNP3LM3KVFFYMO6GPIYRV/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":"2026-07-12T20:12:05Z","links":{"resolver":"https://pith.science/pith/BVB34ZNP3LM3KVFFYMO6GPIYRV","bundle":"https://pith.science/pith/BVB34ZNP3LM3KVFFYMO6GPIYRV/bundle.json","state":"https://pith.science/pith/BVB34ZNP3LM3KVFFYMO6GPIYRV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BVB34ZNP3LM3KVFFYMO6GPIYRV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BVB34ZNP3LM3KVFFYMO6GPIYRV","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":"a310394a3d57bf0021f71b9d2a87042aa64a8eaa2080999776dd84e53d88d585","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-19T13:33:29Z","title_canon_sha256":"7df2764720634f52c55509dc9f51921a4eabec884184655d80de5e10dafb322f"},"schema_version":"1.0","source":{"id":"2503.15195","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.15195","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"arxiv_version","alias_value":"2503.15195v3","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.15195","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_12","alias_value":"BVB34ZNP3LM3","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_16","alias_value":"BVB34ZNP3LM3KVFF","created_at":"2026-07-05T11:25:42Z"},{"alias_kind":"pith_short_8","alias_value":"BVB34ZNP","created_at":"2026-07-05T11:25:42Z"}],"graph_snapshots":[{"event_id":"sha256:8cceff09ca22879e876c377e6f2656e849aaac220819748ff805987810277502","target":"graph","created_at":"2026-07-05T11:25:42Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2503.15195/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Traditional machine learning models for Handwritten Text Recognition (HTR) rely on supervised training, requiring extensive manual annotations, and often produce errors due to the separation between layout and text processing. In contrast, Multimodal Large Language Models (MLLMs) offer a general approach to recognizing diverse handwriting styles without the need for model-specific training. The study benchmarks various proprietary and open-source LLMs against Transkribus models, evaluating their performance on both modern and historical datasets written in English, French, German, and Italian.","authors_text":"Giorgia Crosilla, Giovanni Colavizza, Lukas Klic","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-19T13:33:29Z","title":"Benchmarking Large Language Models for Handwritten Text Recognition"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.15195","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:cf2b9d0b3bbb3860fa15bd78853841e35763cfa6ed746eb96f15bc572ed13482","target":"record","created_at":"2026-07-05T11:25:42Z","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":"a310394a3d57bf0021f71b9d2a87042aa64a8eaa2080999776dd84e53d88d585","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-19T13:33:29Z","title_canon_sha256":"7df2764720634f52c55509dc9f51921a4eabec884184655d80de5e10dafb322f"},"schema_version":"1.0","source":{"id":"2503.15195","kind":"arxiv","version":3}},"canonical_sha256":"0d43be65afdad9b554a5c31de33d188d75515c38caa7344fd7ed0f0d7474c783","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0d43be65afdad9b554a5c31de33d188d75515c38caa7344fd7ed0f0d7474c783","first_computed_at":"2026-07-05T11:25:42.914213Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:25:42.914213Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2UvMsSjSODvunFB1MuH90pS3IR25rj85JqmGu3hZcg7dsa/W28hqn7glB3BWJ806oH7+i0bVVT4EcUEvkfbqAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:25:42.914648Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.15195","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cf2b9d0b3bbb3860fa15bd78853841e35763cfa6ed746eb96f15bc572ed13482","sha256:8cceff09ca22879e876c377e6f2656e849aaac220819748ff805987810277502"],"state_sha256":"84819e15d1939fdebc20f806d9b837bd86f00d1e0503387281624a186e2b1044"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/1kW2ATRb8vY5OVtK6GxC02VJ1fBeRtrCwc0CyQOtnhT9DkVjiR4r/Fu1Pm5tuDRieayHHycbL72bUjGMP9LAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-12T20:12:05.913553Z","bundle_sha256":"d8fd8c6c2467f7bb6febf05422d7f17693bc4338feb9033d62edf55b7e6c97a9"}}