{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:PHV67C645GFZLJFZEFEWRXFIFW","short_pith_number":"pith:PHV67C64","schema_version":"1.0","canonical_sha256":"79ebef8bdce98b95a4b9214968dca82d8ed67d89143b6bcab8e5d77282134793","source":{"kind":"arxiv","id":"2405.14295","version":1},"attestation_state":"computed","paper":{"title":"Focus Anywhere for Fine-grained Multi-page Document Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chenglong Liu, Chunrui Han, Haoran Wei, Jianjian Sun, Jinyue Chen, Liang Zhao, Lingyu Kong, Xiangyu Zhang, Zheng Ge, Zining Zhu","submitted_at":"2024-05-23T08:15:49Z","abstract_excerpt":"Modern LVLMs still struggle to achieve fine-grained document understanding, such as OCR/translation/caption for regions of interest to the user, tasks that require the context of the entire page, or even multiple pages. Accordingly, this paper proposes Fox, an effective pipeline, hybrid data, and tuning strategy, that catalyzes LVLMs to focus anywhere on single/multi-page documents. We introduce a novel task to boost the document understanding by making LVLMs focus attention on the document-level region, such as redefining full-page OCR as foreground focus. We employ multiple vision vocabulari"},"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":"2405.14295","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-05-23T08:15:49Z","cross_cats_sorted":[],"title_canon_sha256":"af232d658d3af2ae0478ffddfc76723351eb86ca92ec78d798873d47afb8f7bb","abstract_canon_sha256":"5a7164bc9eed80a38b7c37db66a3afc994c8c0d67ca704d570f9bad1e6a677f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:22:12.554098Z","signature_b64":"JtCaoc24IoVhmrHd9+yqFs80yibLv38yW0U1SiD8Ad82oavMCj5i0YESu/3k4mmot/rMTSAwPFJzmGh+GJjUCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"79ebef8bdce98b95a4b9214968dca82d8ed67d89143b6bcab8e5d77282134793","last_reissued_at":"2026-07-05T08:22:12.553589Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:22:12.553589Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Focus Anywhere for Fine-grained Multi-page Document Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chenglong Liu, Chunrui Han, Haoran Wei, Jianjian Sun, Jinyue Chen, Liang Zhao, Lingyu Kong, Xiangyu Zhang, Zheng Ge, Zining Zhu","submitted_at":"2024-05-23T08:15:49Z","abstract_excerpt":"Modern LVLMs still struggle to achieve fine-grained document understanding, such as OCR/translation/caption for regions of interest to the user, tasks that require the context of the entire page, or even multiple pages. Accordingly, this paper proposes Fox, an effective pipeline, hybrid data, and tuning strategy, that catalyzes LVLMs to focus anywhere on single/multi-page documents. We introduce a novel task to boost the document understanding by making LVLMs focus attention on the document-level region, such as redefining full-page OCR as foreground focus. We employ multiple vision vocabulari"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14295","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/2405.14295/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":"2405.14295","created_at":"2026-07-05T08:22:12.553652+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.14295v1","created_at":"2026-07-05T08:22:12.553652+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14295","created_at":"2026-07-05T08:22:12.553652+00:00"},{"alias_kind":"pith_short_12","alias_value":"PHV67C645GFZ","created_at":"2026-07-05T08:22:12.553652+00:00"},{"alias_kind":"pith_short_16","alias_value":"PHV67C645GFZLJFZ","created_at":"2026-07-05T08:22:12.553652+00:00"},{"alias_kind":"pith_short_8","alias_value":"PHV67C64","created_at":"2026-07-05T08:22:12.553652+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05927","citing_title":"CMDR: Contextual Multimodal Document Retrieval","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22100","citing_title":"MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2410.05970","citing_title":"PDF-WuKong: A Large Multimodal Model for Efficient Long PDF Reading with End-to-End Sparse Sampling","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2410.21169","citing_title":"Document Parsing Unveiled: Techniques, Challenges, and Prospects for Structured Information Extraction","ref_index":133,"is_internal_anchor":false},{"citing_arxiv_id":"2504.11101","citing_title":"Consensus Entropy: Harnessing Multi-VLM Agreement for Self-Verifying and Self-Improving OCR","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22100","citing_title":"MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2409.01704","citing_title":"General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2501.00321","citing_title":"OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2511.14998","citing_title":"FinCriticalED: A Visual Benchmark for Financial Fact-Level OCR","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2409.18839","citing_title":"MinerU: An Open-Source Solution for Precise Document Content Extraction","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2603.23885","citing_title":"Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware Training","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2510.18234","citing_title":"DeepSeek-OCR: Contexts Optical Compression","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00392","citing_title":"RTPrune: Reading-Twice Inspired Token Pruning for Efficient DeepSeek-OCR Inference","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PHV67C645GFZLJFZEFEWRXFIFW","json":"https://pith.science/pith/PHV67C645GFZLJFZEFEWRXFIFW.json","graph_json":"https://pith.science/api/pith-number/PHV67C645GFZLJFZEFEWRXFIFW/graph.json","events_json":"https://pith.science/api/pith-number/PHV67C645GFZLJFZEFEWRXFIFW/events.json","paper":"https://pith.science/paper/PHV67C64"},"agent_actions":{"view_html":"https://pith.science/pith/PHV67C645GFZLJFZEFEWRXFIFW","download_json":"https://pith.science/pith/PHV67C645GFZLJFZEFEWRXFIFW.json","view_paper":"https://pith.science/paper/PHV67C64","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.14295&json=true","fetch_graph":"https://pith.science/api/pith-number/PHV67C645GFZLJFZEFEWRXFIFW/graph.json","fetch_events":"https://pith.science/api/pith-number/PHV67C645GFZLJFZEFEWRXFIFW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PHV67C645GFZLJFZEFEWRXFIFW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PHV67C645GFZLJFZEFEWRXFIFW/action/storage_attestation","attest_author":"https://pith.science/pith/PHV67C645GFZLJFZEFEWRXFIFW/action/author_attestation","sign_citation":"https://pith.science/pith/PHV67C645GFZLJFZEFEWRXFIFW/action/citation_signature","submit_replication":"https://pith.science/pith/PHV67C645GFZLJFZEFEWRXFIFW/action/replication_record"}},"created_at":"2026-07-05T08:22:12.553652+00:00","updated_at":"2026-07-05T08:22:12.553652+00:00"}