{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KUU7YHX6JCERWDMMOS53C2B4LM","short_pith_number":"pith:KUU7YHX6","schema_version":"1.0","canonical_sha256":"5529fc1efe48891b0d8c74bbb1683c5b3a4c524cda2b668bcb8a755589f3cc62","source":{"kind":"arxiv","id":"2406.19101","version":4},"attestation_state":"computed","paper":{"title":"DocKylin: A Large Multimodal Model for Visual Document Understanding with Efficient Visual Slimming","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiaxin Zhang, Lianwen Jin, Songxuan Lai, Wentao Yang, Zecheng Xie","submitted_at":"2024-06-27T11:28:36Z","abstract_excerpt":"Current multimodal large language models (MLLMs) face significant challenges in visual document understanding (VDU) tasks due to the high resolution, dense text, and complex layouts typical of document images. These characteristics demand a high level of detail perception ability from MLLMs. While increasing input resolution improves detail perception capability, it also leads to longer sequences of visual tokens, increasing computational costs and straining the models' ability to handle long contexts. To address these challenges, we introduce DocKylin, a document-centric MLLM that performs vi"},"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":"2406.19101","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-06-27T11:28:36Z","cross_cats_sorted":[],"title_canon_sha256":"c26f3929865fe3fc60a9cb3f387c76862dbc0f7814cb73ab9ab25436c15b7410","abstract_canon_sha256":"2d873f79e63c02310dceb3fd8c6b75d4c9dfa4e65074be30ff8d1773a97f3791"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:51:34.000228Z","signature_b64":"fD2JRzMgRx2DZa9HO5FIBVepCSc4J31jigbGpVcYg8omXN0dS2dI7pa7TeWXnSzGl012RP74f0qzhqqSEM0fCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5529fc1efe48891b0d8c74bbb1683c5b3a4c524cda2b668bcb8a755589f3cc62","last_reissued_at":"2026-07-05T09:51:33.999805Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:51:33.999805Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DocKylin: A Large Multimodal Model for Visual Document Understanding with Efficient Visual Slimming","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiaxin Zhang, Lianwen Jin, Songxuan Lai, Wentao Yang, Zecheng Xie","submitted_at":"2024-06-27T11:28:36Z","abstract_excerpt":"Current multimodal large language models (MLLMs) face significant challenges in visual document understanding (VDU) tasks due to the high resolution, dense text, and complex layouts typical of document images. These characteristics demand a high level of detail perception ability from MLLMs. While increasing input resolution improves detail perception capability, it also leads to longer sequences of visual tokens, increasing computational costs and straining the models' ability to handle long contexts. To address these challenges, we introduce DocKylin, a document-centric MLLM that performs vi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.19101","kind":"arxiv","version":4},"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/2406.19101/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":"2406.19101","created_at":"2026-07-05T09:51:33.999866+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.19101v4","created_at":"2026-07-05T09:51:33.999866+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.19101","created_at":"2026-07-05T09:51:33.999866+00:00"},{"alias_kind":"pith_short_12","alias_value":"KUU7YHX6JCER","created_at":"2026-07-05T09:51:33.999866+00:00"},{"alias_kind":"pith_short_16","alias_value":"KUU7YHX6JCERWDMM","created_at":"2026-07-05T09:51:33.999866+00:00"},{"alias_kind":"pith_short_8","alias_value":"KUU7YHX6","created_at":"2026-07-05T09:51:33.999866+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2507.09861","citing_title":"A Survey on MLLM-based Visually Rich Document Understanding: Methods, Challenges, and Emerging Trends","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2604.00161","citing_title":"Q-Mask: Query-driven Causal Masks for Text Anchoring in OCR-Oriented Vision-Language Models","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KUU7YHX6JCERWDMMOS53C2B4LM","json":"https://pith.science/pith/KUU7YHX6JCERWDMMOS53C2B4LM.json","graph_json":"https://pith.science/api/pith-number/KUU7YHX6JCERWDMMOS53C2B4LM/graph.json","events_json":"https://pith.science/api/pith-number/KUU7YHX6JCERWDMMOS53C2B4LM/events.json","paper":"https://pith.science/paper/KUU7YHX6"},"agent_actions":{"view_html":"https://pith.science/pith/KUU7YHX6JCERWDMMOS53C2B4LM","download_json":"https://pith.science/pith/KUU7YHX6JCERWDMMOS53C2B4LM.json","view_paper":"https://pith.science/paper/KUU7YHX6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.19101&json=true","fetch_graph":"https://pith.science/api/pith-number/KUU7YHX6JCERWDMMOS53C2B4LM/graph.json","fetch_events":"https://pith.science/api/pith-number/KUU7YHX6JCERWDMMOS53C2B4LM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KUU7YHX6JCERWDMMOS53C2B4LM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KUU7YHX6JCERWDMMOS53C2B4LM/action/storage_attestation","attest_author":"https://pith.science/pith/KUU7YHX6JCERWDMMOS53C2B4LM/action/author_attestation","sign_citation":"https://pith.science/pith/KUU7YHX6JCERWDMMOS53C2B4LM/action/citation_signature","submit_replication":"https://pith.science/pith/KUU7YHX6JCERWDMMOS53C2B4LM/action/replication_record"}},"created_at":"2026-07-05T09:51:33.999866+00:00","updated_at":"2026-07-05T09:51:33.999866+00:00"}