{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:DTL4V2DFPQ7GLCA7CXTP52D4PE","short_pith_number":"pith:DTL4V2DF","schema_version":"1.0","canonical_sha256":"1cd7cae8657c3e65881f15e6fee87c792c82aa1861a175384d0d7cc822801935","source":{"kind":"arxiv","id":"2406.07230","version":2},"attestation_state":"computed","paper":{"title":"Needle In A Multimodal Haystack","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jifeng Dai, Kaipeng Zhang, Lewei Lu, Mengkang Hu, Ping Luo, Shuibo Zhang, Shuo Liu, Tiantong Li, Weiyun Wang, Wenhai Wang, Wenqi Shao, Xizhou Zhu, Yiming Ren, Yuchen Duan, Yu Qiao, Zhe Chen","submitted_at":"2024-06-11T13:09:16Z","abstract_excerpt":"With the rapid advancement of multimodal large language models (MLLMs), their evaluation has become increasingly comprehensive. However, understanding long multimodal content, as a foundational ability for real-world applications, remains underexplored. In this work, we present Needle In A Multimodal Haystack (MM-NIAH), the first benchmark specifically designed to systematically evaluate the capability of existing MLLMs to comprehend long multimodal documents. Our benchmark includes three types of evaluation tasks: multimodal retrieval, counting, and reasoning. In each task, the model is requi"},"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":"2406.07230","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-06-11T13:09:16Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a5c6ea84d2e31fd76ad1e66a864ff8f66a7257e6f884bccc9269e9668ac0b5f6","abstract_canon_sha256":"8746613dec7de29f987707178b991d385504f4dd14411a805be4f16b46a902f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:18:27.047594Z","signature_b64":"arjRuhn0qskDLtIT+wmmv5MJgweUxN1OhG4+Orl1gbbOPxzgNlDC6ipmnFfqoPkqV56B6mIWFuuDba2taLmXDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1cd7cae8657c3e65881f15e6fee87c792c82aa1861a175384d0d7cc822801935","last_reissued_at":"2026-07-05T09:18:27.047113Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:18:27.047113Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Needle In A Multimodal Haystack","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jifeng Dai, Kaipeng Zhang, Lewei Lu, Mengkang Hu, Ping Luo, Shuibo Zhang, Shuo Liu, Tiantong Li, Weiyun Wang, Wenhai Wang, Wenqi Shao, Xizhou Zhu, Yiming Ren, Yuchen Duan, Yu Qiao, Zhe Chen","submitted_at":"2024-06-11T13:09:16Z","abstract_excerpt":"With the rapid advancement of multimodal large language models (MLLMs), their evaluation has become increasingly comprehensive. However, understanding long multimodal content, as a foundational ability for real-world applications, remains underexplored. In this work, we present Needle In A Multimodal Haystack (MM-NIAH), the first benchmark specifically designed to systematically evaluate the capability of existing MLLMs to comprehend long multimodal documents. Our benchmark includes three types of evaluation tasks: multimodal retrieval, counting, and reasoning. In each task, the model is requi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.07230","kind":"arxiv","version":2},"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/2406.07230/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":"2406.07230","created_at":"2026-07-05T09:18:27.047164+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.07230v2","created_at":"2026-07-05T09:18:27.047164+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.07230","created_at":"2026-07-05T09:18:27.047164+00:00"},{"alias_kind":"pith_short_12","alias_value":"DTL4V2DFPQ7G","created_at":"2026-07-05T09:18:27.047164+00:00"},{"alias_kind":"pith_short_16","alias_value":"DTL4V2DFPQ7GLCA7","created_at":"2026-07-05T09:18:27.047164+00:00"},{"alias_kind":"pith_short_8","alias_value":"DTL4V2DF","created_at":"2026-07-05T09:18:27.047164+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14906","citing_title":"MemLens: Benchmarking Multimodal Long-Term Memory in Large Vision-Language Models","ref_index":12,"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":18,"is_internal_anchor":false},{"citing_arxiv_id":"2408.04840","citing_title":"mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models","ref_index":251,"is_internal_anchor":false},{"citing_arxiv_id":"2411.10442","citing_title":"Enhancing the Reasoning Ability of Multimodal Large Language Models via Mixed Preference Optimization","ref_index":103,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DTL4V2DFPQ7GLCA7CXTP52D4PE","json":"https://pith.science/pith/DTL4V2DFPQ7GLCA7CXTP52D4PE.json","graph_json":"https://pith.science/api/pith-number/DTL4V2DFPQ7GLCA7CXTP52D4PE/graph.json","events_json":"https://pith.science/api/pith-number/DTL4V2DFPQ7GLCA7CXTP52D4PE/events.json","paper":"https://pith.science/paper/DTL4V2DF"},"agent_actions":{"view_html":"https://pith.science/pith/DTL4V2DFPQ7GLCA7CXTP52D4PE","download_json":"https://pith.science/pith/DTL4V2DFPQ7GLCA7CXTP52D4PE.json","view_paper":"https://pith.science/paper/DTL4V2DF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.07230&json=true","fetch_graph":"https://pith.science/api/pith-number/DTL4V2DFPQ7GLCA7CXTP52D4PE/graph.json","fetch_events":"https://pith.science/api/pith-number/DTL4V2DFPQ7GLCA7CXTP52D4PE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DTL4V2DFPQ7GLCA7CXTP52D4PE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DTL4V2DFPQ7GLCA7CXTP52D4PE/action/storage_attestation","attest_author":"https://pith.science/pith/DTL4V2DFPQ7GLCA7CXTP52D4PE/action/author_attestation","sign_citation":"https://pith.science/pith/DTL4V2DFPQ7GLCA7CXTP52D4PE/action/citation_signature","submit_replication":"https://pith.science/pith/DTL4V2DFPQ7GLCA7CXTP52D4PE/action/replication_record"}},"created_at":"2026-07-05T09:18:27.047164+00:00","updated_at":"2026-07-05T09:18:27.047164+00:00"}