{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:33CWAPAZQKEK3FJTRCWUWYYTNG","short_pith_number":"pith:33CWAPAZ","schema_version":"1.0","canonical_sha256":"dec5603c198288ad953388ad4b63136993a6cf002029d285a6bccc41af23fcc2","source":{"kind":"arxiv","id":"2407.01523","version":3},"attestation_state":"computed","paper":{"title":"MMLongBench-Doc: Benchmarking Long-context Document Understanding with Visualizations","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Aixin Sun, Jiaqi Wang, Liangming Pan, Liangyu Chen, Meiqi Chen, Pan Zhang, Xiaoyi Dong, Xinyuan Lu, Xinze Li, Yan Ma, Yixin Cao, Yizhu Jiao, Yubo Ma, Yu-Gang Jiang, Yuhang Zang, Ziyu Liu","submitted_at":"2024-07-01T17:59:26Z","abstract_excerpt":"Understanding documents with rich layouts and multi-modal components is a long-standing and practical task. Recent Large Vision-Language Models (LVLMs) have made remarkable strides in various tasks, particularly in single-page document understanding (DU). However, their abilities on long-context DU remain an open problem. This work presents MMLongBench-Doc, a long-context, multi-modal benchmark comprising 1,062 expert-annotated questions. Distinct from previous datasets, it is constructed upon 130 lengthy PDF-formatted documents with an average of 49.4 pages and 20,971 textual tokens. Towards "},"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":"2407.01523","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2024-07-01T17:59:26Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"c8a1ce314e7298a120facda62c7b977a5521372c36db15b07ac96f96de9c8817","abstract_canon_sha256":"ec782e52caaf8fe682dc54af4ba46b190fa201b266507c891b4bbc9c7562806c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:34:15.468749Z","signature_b64":"0RToxJVbxka2FumJPrwX2g3ITn8gOEL1s5SCeZnuvaC0M3vTpiaaX9cZcKhGDbe/2/w7D9R1HACL0jxDCuXgCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dec5603c198288ad953388ad4b63136993a6cf002029d285a6bccc41af23fcc2","last_reissued_at":"2026-07-05T09:34:15.468249Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:34:15.468249Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MMLongBench-Doc: Benchmarking Long-context Document Understanding with Visualizations","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Aixin Sun, Jiaqi Wang, Liangming Pan, Liangyu Chen, Meiqi Chen, Pan Zhang, Xiaoyi Dong, Xinyuan Lu, Xinze Li, Yan Ma, Yixin Cao, Yizhu Jiao, Yubo Ma, Yu-Gang Jiang, Yuhang Zang, Ziyu Liu","submitted_at":"2024-07-01T17:59:26Z","abstract_excerpt":"Understanding documents with rich layouts and multi-modal components is a long-standing and practical task. Recent Large Vision-Language Models (LVLMs) have made remarkable strides in various tasks, particularly in single-page document understanding (DU). However, their abilities on long-context DU remain an open problem. This work presents MMLongBench-Doc, a long-context, multi-modal benchmark comprising 1,062 expert-annotated questions. Distinct from previous datasets, it is constructed upon 130 lengthy PDF-formatted documents with an average of 49.4 pages and 20,971 textual tokens. Towards "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.01523","kind":"arxiv","version":3},"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/2407.01523/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":"2407.01523","created_at":"2026-07-05T09:34:15.468309+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.01523v3","created_at":"2026-07-05T09:34:15.468309+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.01523","created_at":"2026-07-05T09:34:15.468309+00:00"},{"alias_kind":"pith_short_12","alias_value":"33CWAPAZQKEK","created_at":"2026-07-05T09:34:15.468309+00:00"},{"alias_kind":"pith_short_16","alias_value":"33CWAPAZQKEK3FJT","created_at":"2026-07-05T09:34:15.468309+00:00"},{"alias_kind":"pith_short_8","alias_value":"33CWAPAZ","created_at":"2026-07-05T09:34:15.468309+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07703","citing_title":"How Much Dense Attention is Necessary? Oracle-Guided Sparse Prefill for Full/GQA Layers in Hybrid Long-Context Models","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14906","citing_title":"MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29399","citing_title":"LLM-Guided Planning for Multi-hop Reasoning over Multimodal Nuclear Regulatory Documents","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29648","citing_title":"Hybrid Retriever Evolution for Multimodal Document Reasoning Agents","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30027","citing_title":"DocRetriever: A Plug-and-Play Framework for Multimodal Document Retrieval with Comprehensive Benchmark","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2510.18830","citing_title":"MTraining: Distributed Dynamic Sparse Attention for Efficient Ultra-Long Context Training","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2507.04590","citing_title":"VLM2Vec-V2: Advancing Multimodal Embedding for Videos, Images, and Visual Documents","ref_index":18,"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":27,"is_internal_anchor":false},{"citing_arxiv_id":"2407.03320","citing_title":"InternLM-XComposer-2.5: A Versatile Large Vision Language Model Supporting Long-Contextual Input and Output","ref_index":104,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02371","citing_title":"Internalized Reasoning for Long-Context Visual Document Understanding","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26837","citing_title":"Unifying Sparse Attention with Hierarchical Memory for Scalable Long-Context LLM Serving","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04901","citing_title":"FileGram: Grounding Agent Personalization in File-System Behavioral Traces","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/33CWAPAZQKEK3FJTRCWUWYYTNG","json":"https://pith.science/pith/33CWAPAZQKEK3FJTRCWUWYYTNG.json","graph_json":"https://pith.science/api/pith-number/33CWAPAZQKEK3FJTRCWUWYYTNG/graph.json","events_json":"https://pith.science/api/pith-number/33CWAPAZQKEK3FJTRCWUWYYTNG/events.json","paper":"https://pith.science/paper/33CWAPAZ"},"agent_actions":{"view_html":"https://pith.science/pith/33CWAPAZQKEK3FJTRCWUWYYTNG","download_json":"https://pith.science/pith/33CWAPAZQKEK3FJTRCWUWYYTNG.json","view_paper":"https://pith.science/paper/33CWAPAZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.01523&json=true","fetch_graph":"https://pith.science/api/pith-number/33CWAPAZQKEK3FJTRCWUWYYTNG/graph.json","fetch_events":"https://pith.science/api/pith-number/33CWAPAZQKEK3FJTRCWUWYYTNG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/33CWAPAZQKEK3FJTRCWUWYYTNG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/33CWAPAZQKEK3FJTRCWUWYYTNG/action/storage_attestation","attest_author":"https://pith.science/pith/33CWAPAZQKEK3FJTRCWUWYYTNG/action/author_attestation","sign_citation":"https://pith.science/pith/33CWAPAZQKEK3FJTRCWUWYYTNG/action/citation_signature","submit_replication":"https://pith.science/pith/33CWAPAZQKEK3FJTRCWUWYYTNG/action/replication_record"}},"created_at":"2026-07-05T09:34:15.468309+00:00","updated_at":"2026-07-05T09:34:15.468309+00:00"}