{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:IES4XS457CXNKWFOA5NQITCDL4","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":"96eeafc663b24976c99d778b7e4d58be62dfdc59f3988ba38d6d81775206dbb3","cross_cats_sorted":["cs.CV","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2022-09-26T10:27:29Z","title_canon_sha256":"66e62447d1e89be1126be0a78df0ab0365e0eacaf289a080699b987aeba83216"},"schema_version":"1.0","source":{"id":"2209.12561","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2209.12561","created_at":"2026-07-05T05:00:53Z"},{"alias_kind":"arxiv_version","alias_value":"2209.12561v1","created_at":"2026-07-05T05:00:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.12561","created_at":"2026-07-05T05:00:53Z"},{"alias_kind":"pith_short_12","alias_value":"IES4XS457CXN","created_at":"2026-07-05T05:00:53Z"},{"alias_kind":"pith_short_16","alias_value":"IES4XS457CXNKWFO","created_at":"2026-07-05T05:00:53Z"},{"alias_kind":"pith_short_8","alias_value":"IES4XS45","created_at":"2026-07-05T05:00:53Z"}],"graph_snapshots":[{"event_id":"sha256:b6d19f018c99ef7698736f2c18c0c83e84a23869ef0a173f7a379163b51b59ef","target":"graph","created_at":"2026-07-05T05:00:53Z","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/2209.12561/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Successful Artificial Intelligence systems often require numerous labeled data to extract information from document images. In this paper, we investigate the problem of improving the performance of Artificial Intelligence systems in understanding document images, especially in cases where training data is limited. We address the problem by proposing a novel finetuning method using reinforcement learning. Our approach treats the Information Extraction model as a policy network and uses policy gradient training to update the model to maximize combined reward functions that complement the traditi","authors_text":"Bao-Sinh Nguyen, Dung Tien Le, Hieu M. Vu, Hung Le, Minh-Tien Nguyen, Tuan Anh D. Nguyen","cross_cats":["cs.CV","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2022-09-26T10:27:29Z","title":"Improving Document Image Understanding with Reinforcement Finetuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.12561","kind":"arxiv","version":1},"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":"sha256:6eaf5520cf092db56da93c9daeec7e9776b5212b6a6af8f7eb512bf79b15f723","target":"record","created_at":"2026-07-05T05:00:53Z","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":"96eeafc663b24976c99d778b7e4d58be62dfdc59f3988ba38d6d81775206dbb3","cross_cats_sorted":["cs.CV","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.IR","submitted_at":"2022-09-26T10:27:29Z","title_canon_sha256":"66e62447d1e89be1126be0a78df0ab0365e0eacaf289a080699b987aeba83216"},"schema_version":"1.0","source":{"id":"2209.12561","kind":"arxiv","version":1}},"canonical_sha256":"4125cbcb9df8aed558ae075b044c435f21b9c44ec6815635108f90877f33046f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4125cbcb9df8aed558ae075b044c435f21b9c44ec6815635108f90877f33046f","first_computed_at":"2026-07-05T05:00:53.767539Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:00:53.767539Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9xrqxLuZZqPi16bP2dz9tTBwfwE2eG8X7IkCPJcDaNBQT+8heGbihoJuLQxiLFqwvBRAmlG9sm1D4Ze1at20Cw==","signature_status":"signed_v1","signed_at":"2026-07-05T05:00:53.767950Z","signed_message":"canonical_sha256_bytes"},"source_id":"2209.12561","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6eaf5520cf092db56da93c9daeec7e9776b5212b6a6af8f7eb512bf79b15f723","sha256:b6d19f018c99ef7698736f2c18c0c83e84a23869ef0a173f7a379163b51b59ef"],"state_sha256":"2ba78cb722a4850e67e18022b055f05e7923f90f795a52f7eeab85e8abb01561"}