{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:XRC2AQVAM5TGE5CTGLR7CBEDKS","short_pith_number":"pith:XRC2AQVA","schema_version":"1.0","canonical_sha256":"bc45a042a0676662745332e3f10483548f7258efc4ea4911bc440d8b4cc7b2c3","source":{"kind":"arxiv","id":"2404.09204","version":1},"attestation_state":"computed","paper":{"title":"TextHawk: Exploring Efficient Fine-Grained Perception of Multimodal Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jihao Wu, Minghui Liao, Wei Zeng, Xiaoyu Zheng, Ya-Qi Yu, Yongxin Liao","submitted_at":"2024-04-14T09:48:37Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) have shown impressive results on various multimodal tasks. However, most existing MLLMs are not well suited for document-oriented tasks, which require fine-grained image perception and information compression. In this paper, we present TextHawk, a MLLM that is specifically designed for document-oriented tasks, while preserving the general capabilities of MLLMs. TextHawk is aimed to explore efficient fine-grained perception by designing four dedicated components. Firstly, a ReSampling and ReArrangement (ReSA) module is proposed to reduce the redundancy i"},"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":"2404.09204","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-04-14T09:48:37Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"7f46a705ab2db43653937ba5fcfb1ddf7a29ecc999d609881200711e2187a432","abstract_canon_sha256":"debb07c642c3c0b5e526de83fc718966c1894e1b777a4a4929ea92ecea54e77f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:07:46.742808Z","signature_b64":"q4V193Ln+h6/J4TMXmzVGOukzGtDpyZOhMlZd5DrE7vKBWOzKSGjEbrmnxt9z+2YrYsU1hSL6uJUVCs+PdyLBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc45a042a0676662745332e3f10483548f7258efc4ea4911bc440d8b4cc7b2c3","last_reissued_at":"2026-07-05T08:07:46.742259Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:07:46.742259Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TextHawk: Exploring Efficient Fine-Grained Perception of Multimodal Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jihao Wu, Minghui Liao, Wei Zeng, Xiaoyu Zheng, Ya-Qi Yu, Yongxin Liao","submitted_at":"2024-04-14T09:48:37Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) have shown impressive results on various multimodal tasks. However, most existing MLLMs are not well suited for document-oriented tasks, which require fine-grained image perception and information compression. In this paper, we present TextHawk, a MLLM that is specifically designed for document-oriented tasks, while preserving the general capabilities of MLLMs. TextHawk is aimed to explore efficient fine-grained perception by designing four dedicated components. Firstly, a ReSampling and ReArrangement (ReSA) module is proposed to reduce the redundancy i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.09204","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/2404.09204/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":"2404.09204","created_at":"2026-07-05T08:07:46.742325+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.09204v1","created_at":"2026-07-05T08:07:46.742325+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.09204","created_at":"2026-07-05T08:07:46.742325+00:00"},{"alias_kind":"pith_short_12","alias_value":"XRC2AQVAM5TG","created_at":"2026-07-05T08:07:46.742325+00:00"},{"alias_kind":"pith_short_16","alias_value":"XRC2AQVAM5TGE5CT","created_at":"2026-07-05T08:07:46.742325+00:00"},{"alias_kind":"pith_short_8","alias_value":"XRC2AQVA","created_at":"2026-07-05T08:07:46.742325+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.30307","citing_title":"Grounded 3D-Aware Spatial Vision-Language Modeling","ref_index":119,"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":62,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19307","citing_title":"MetaRA: Metamorphic Robustness Assessment for Multimodal Large Language Model-based Visual Question Answering Systems","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2507.09861","citing_title":"A Survey on MLLM-based Visually Rich Document Understanding: Methods, Challenges, and Emerging Trends","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2603.18472","citing_title":"Cognitive Mismatch in Multimodal Large Language Models for Discrete Symbol Understanding","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2603.19790","citing_title":"From Plausibility to Verifiability: Risk-Controlled Generative OCR with Vision-Language Models","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07419","citing_title":"ReAlign: Optimizing the Visual Document Retriever with Reasoning-Guided Fine-Grained Alignment","ref_index":95,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XRC2AQVAM5TGE5CTGLR7CBEDKS","json":"https://pith.science/pith/XRC2AQVAM5TGE5CTGLR7CBEDKS.json","graph_json":"https://pith.science/api/pith-number/XRC2AQVAM5TGE5CTGLR7CBEDKS/graph.json","events_json":"https://pith.science/api/pith-number/XRC2AQVAM5TGE5CTGLR7CBEDKS/events.json","paper":"https://pith.science/paper/XRC2AQVA"},"agent_actions":{"view_html":"https://pith.science/pith/XRC2AQVAM5TGE5CTGLR7CBEDKS","download_json":"https://pith.science/pith/XRC2AQVAM5TGE5CTGLR7CBEDKS.json","view_paper":"https://pith.science/paper/XRC2AQVA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.09204&json=true","fetch_graph":"https://pith.science/api/pith-number/XRC2AQVAM5TGE5CTGLR7CBEDKS/graph.json","fetch_events":"https://pith.science/api/pith-number/XRC2AQVAM5TGE5CTGLR7CBEDKS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XRC2AQVAM5TGE5CTGLR7CBEDKS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XRC2AQVAM5TGE5CTGLR7CBEDKS/action/storage_attestation","attest_author":"https://pith.science/pith/XRC2AQVAM5TGE5CTGLR7CBEDKS/action/author_attestation","sign_citation":"https://pith.science/pith/XRC2AQVAM5TGE5CTGLR7CBEDKS/action/citation_signature","submit_replication":"https://pith.science/pith/XRC2AQVAM5TGE5CTGLR7CBEDKS/action/replication_record"}},"created_at":"2026-07-05T08:07:46.742325+00:00","updated_at":"2026-07-05T08:07:46.742325+00:00"}