{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:62MP56XOSQKDTXSEYXQ5LO6NWW","short_pith_number":"pith:62MP56XO","schema_version":"1.0","canonical_sha256":"f698fefaee941439de44c5e1d5bbcdb5aa891414db6db5ef0b3e431f05af681c","source":{"kind":"arxiv","id":"2507.09531","version":1},"attestation_state":"computed","paper":{"title":"VDInstruct: Zero-Shot Key Information Extraction via Content-Aware Vision Tokenization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Daeyoung Kim, Giang Nguyen, Hung Dao, Son Nguyen, Thao Do","submitted_at":"2025-07-13T08:15:11Z","abstract_excerpt":"Key Information Extraction (KIE) underpins the understanding of visual documents (e.g., receipts and contracts) by extracting precise semantic content and accurately capturing spatial structure. Yet existing multimodal large language models (MLLMs) often perform poorly on dense documents and rely on vision tokenization approaches that scale with image size, leading to redundant computation and memory inefficiency. To address these challenges, we introduce VDInstruct, an MLLM that separates spatial region detection from semantic feature extraction. Central to our model is a content-aware tokeni"},"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":"2507.09531","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-07-13T08:15:11Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"4ab675741acbda4c57584cfed69fedffc54fe5b1ef3569db8ce0a312f08873fc","abstract_canon_sha256":"7b89c56d7cd8f18c834e04b9a84636d4ce48f1a7cead4580172516fcf789e6c9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:36:10.688596Z","signature_b64":"vFctNsN87xUbrWQYKJx1B53K93VqBISuHr0m+p1CzqHMjt/4UFQJJQswhJnfywhxkgVg/l5UXVM4ofRGs6sxBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f698fefaee941439de44c5e1d5bbcdb5aa891414db6db5ef0b3e431f05af681c","last_reissued_at":"2026-07-05T11:36:10.688134Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:36:10.688134Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"VDInstruct: Zero-Shot Key Information Extraction via Content-Aware Vision Tokenization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Daeyoung Kim, Giang Nguyen, Hung Dao, Son Nguyen, Thao Do","submitted_at":"2025-07-13T08:15:11Z","abstract_excerpt":"Key Information Extraction (KIE) underpins the understanding of visual documents (e.g., receipts and contracts) by extracting precise semantic content and accurately capturing spatial structure. Yet existing multimodal large language models (MLLMs) often perform poorly on dense documents and rely on vision tokenization approaches that scale with image size, leading to redundant computation and memory inefficiency. To address these challenges, we introduce VDInstruct, an MLLM that separates spatial region detection from semantic feature extraction. Central to our model is a content-aware tokeni"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.09531","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/2507.09531/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":"2507.09531","created_at":"2026-07-05T11:36:10.688198+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.09531v1","created_at":"2026-07-05T11:36:10.688198+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.09531","created_at":"2026-07-05T11:36:10.688198+00:00"},{"alias_kind":"pith_short_12","alias_value":"62MP56XOSQKD","created_at":"2026-07-05T11:36:10.688198+00:00"},{"alias_kind":"pith_short_16","alias_value":"62MP56XOSQKDTXSE","created_at":"2026-07-05T11:36:10.688198+00:00"},{"alias_kind":"pith_short_8","alias_value":"62MP56XO","created_at":"2026-07-05T11:36:10.688198+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/62MP56XOSQKDTXSEYXQ5LO6NWW","json":"https://pith.science/pith/62MP56XOSQKDTXSEYXQ5LO6NWW.json","graph_json":"https://pith.science/api/pith-number/62MP56XOSQKDTXSEYXQ5LO6NWW/graph.json","events_json":"https://pith.science/api/pith-number/62MP56XOSQKDTXSEYXQ5LO6NWW/events.json","paper":"https://pith.science/paper/62MP56XO"},"agent_actions":{"view_html":"https://pith.science/pith/62MP56XOSQKDTXSEYXQ5LO6NWW","download_json":"https://pith.science/pith/62MP56XOSQKDTXSEYXQ5LO6NWW.json","view_paper":"https://pith.science/paper/62MP56XO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.09531&json=true","fetch_graph":"https://pith.science/api/pith-number/62MP56XOSQKDTXSEYXQ5LO6NWW/graph.json","fetch_events":"https://pith.science/api/pith-number/62MP56XOSQKDTXSEYXQ5LO6NWW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/62MP56XOSQKDTXSEYXQ5LO6NWW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/62MP56XOSQKDTXSEYXQ5LO6NWW/action/storage_attestation","attest_author":"https://pith.science/pith/62MP56XOSQKDTXSEYXQ5LO6NWW/action/author_attestation","sign_citation":"https://pith.science/pith/62MP56XOSQKDTXSEYXQ5LO6NWW/action/citation_signature","submit_replication":"https://pith.science/pith/62MP56XOSQKDTXSEYXQ5LO6NWW/action/replication_record"}},"created_at":"2026-07-05T11:36:10.688198+00:00","updated_at":"2026-07-05T11:36:10.688198+00:00"}