{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:F7NU6YHU2S5BTY7TB6OR5CQV7P","short_pith_number":"pith:F7NU6YHU","schema_version":"1.0","canonical_sha256":"2fdb4f60f4d4ba19e3f30f9d1e8a15fbd124c59abf7574c4ba53a18ef5f54e87","source":{"kind":"arxiv","id":"2407.13766","version":4},"attestation_state":"computed","paper":{"title":"Visual Haystacks: A Vision-Centric Needle-In-A-Haystack Benchmark","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"David M. Chan, Giscard Biamby, Jerome Quenum, Joseph E. Gonzalez, Ritwik Gupta, Trevor Darrell, Tsung-Han Wu","submitted_at":"2024-07-18T17:59:30Z","abstract_excerpt":"Large Multimodal Models (LMMs) have made significant strides in visual question-answering for single images. Recent advancements like long-context LMMs have allowed them to ingest larger, or even multiple, images. However, the ability to process a large number of visual tokens does not guarantee effective retrieval and reasoning for multi-image question answering (MIQA), especially in real-world applications like photo album searches or satellite imagery analysis. In this work, we first assess the limitations of current benchmarks for long-context LMMs. We address these limitations by introduc"},"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":"2407.13766","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-07-18T17:59:30Z","cross_cats_sorted":[],"title_canon_sha256":"bfe52e1bb2fe8a74642c3794c994db518ba0e2e1f4a8f6cffbf98a0eaee2c837","abstract_canon_sha256":"eebda950b4a660dd8c87cf638b7b64bdf2ad08e29ff31f6708ee3739354c7146"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:28:19.289742Z","signature_b64":"NB/ftaSpoepHyk8nGjFiMColYnOe3WXfGS79YEdhYuvnLH2fMl5dtaZyHBe8Z12/DJkLK67I6h8eHXuRi7XpAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2fdb4f60f4d4ba19e3f30f9d1e8a15fbd124c59abf7574c4ba53a18ef5f54e87","last_reissued_at":"2026-07-05T10:28:19.289245Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:28:19.289245Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Visual Haystacks: A Vision-Centric Needle-In-A-Haystack Benchmark","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"David M. Chan, Giscard Biamby, Jerome Quenum, Joseph E. Gonzalez, Ritwik Gupta, Trevor Darrell, Tsung-Han Wu","submitted_at":"2024-07-18T17:59:30Z","abstract_excerpt":"Large Multimodal Models (LMMs) have made significant strides in visual question-answering for single images. Recent advancements like long-context LMMs have allowed them to ingest larger, or even multiple, images. However, the ability to process a large number of visual tokens does not guarantee effective retrieval and reasoning for multi-image question answering (MIQA), especially in real-world applications like photo album searches or satellite imagery analysis. In this work, we first assess the limitations of current benchmarks for long-context LMMs. We address these limitations by introduc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.13766","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/2407.13766/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":"2407.13766","created_at":"2026-07-05T10:28:19.289304+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.13766v4","created_at":"2026-07-05T10:28:19.289304+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.13766","created_at":"2026-07-05T10:28:19.289304+00:00"},{"alias_kind":"pith_short_12","alias_value":"F7NU6YHU2S5B","created_at":"2026-07-05T10:28:19.289304+00:00"},{"alias_kind":"pith_short_16","alias_value":"F7NU6YHU2S5BTY7T","created_at":"2026-07-05T10:28:19.289304+00:00"},{"alias_kind":"pith_short_8","alias_value":"F7NU6YHU","created_at":"2026-07-05T10:28:19.289304+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08497","citing_title":"Cognitive-structured Multimodal Agent for Multimodal Understanding, Generation, and Editing","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2606.25445","citing_title":"C3-Bench: A Context-Aware Change Captioning Benchmark","ref_index":93,"is_internal_anchor":false},{"citing_arxiv_id":"2606.25343","citing_title":"Invoice Haystack: Benchmarking Document Retrieval and Visual Question Answering Under Strong Visual Homogeneity","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2606.25343","citing_title":"Invoice Haystack: Benchmarking Document Retrieval and Visual Question Answering Under Strong Visual Homogeneity","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2605.28806","citing_title":"Personal Visual Memory from Explicit and Implicit Evidence","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2603.04676","citing_title":"Decoding the Pulse of Reasoning VLMs in Multi-Image Understanding Tasks","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13831","citing_title":"Training Long-Context Vision-Language Models Effectively with Generalization Beyond 128K Context","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13277","citing_title":"Utility-Oriented Visual Evidence Selection for Multimodal Retrieval-Augmented Generation","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F7NU6YHU2S5BTY7TB6OR5CQV7P","json":"https://pith.science/pith/F7NU6YHU2S5BTY7TB6OR5CQV7P.json","graph_json":"https://pith.science/api/pith-number/F7NU6YHU2S5BTY7TB6OR5CQV7P/graph.json","events_json":"https://pith.science/api/pith-number/F7NU6YHU2S5BTY7TB6OR5CQV7P/events.json","paper":"https://pith.science/paper/F7NU6YHU"},"agent_actions":{"view_html":"https://pith.science/pith/F7NU6YHU2S5BTY7TB6OR5CQV7P","download_json":"https://pith.science/pith/F7NU6YHU2S5BTY7TB6OR5CQV7P.json","view_paper":"https://pith.science/paper/F7NU6YHU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.13766&json=true","fetch_graph":"https://pith.science/api/pith-number/F7NU6YHU2S5BTY7TB6OR5CQV7P/graph.json","fetch_events":"https://pith.science/api/pith-number/F7NU6YHU2S5BTY7TB6OR5CQV7P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F7NU6YHU2S5BTY7TB6OR5CQV7P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F7NU6YHU2S5BTY7TB6OR5CQV7P/action/storage_attestation","attest_author":"https://pith.science/pith/F7NU6YHU2S5BTY7TB6OR5CQV7P/action/author_attestation","sign_citation":"https://pith.science/pith/F7NU6YHU2S5BTY7TB6OR5CQV7P/action/citation_signature","submit_replication":"https://pith.science/pith/F7NU6YHU2S5BTY7TB6OR5CQV7P/action/replication_record"}},"created_at":"2026-07-05T10:28:19.289304+00:00","updated_at":"2026-07-05T10:28:19.289304+00:00"}