{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EUK7U6NQGDREVJJXNEHNK4WVGS","short_pith_number":"pith:EUK7U6NQ","schema_version":"1.0","canonical_sha256":"2515fa79b030e24aa537690ed572d534959f751abd449362a66478b907b77c50","source":{"kind":"arxiv","id":"2408.04840","version":2},"attestation_state":"computed","paper":{"title":"mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Anwen Hu, Fei Huang, Haiyang Xu, Haowei Liu, Jiabo Ye, Jingren Zhou, Ji Zhang, Ming Yan, Qi Qian","submitted_at":"2024-08-09T03:25:42Z","abstract_excerpt":"Multi-modal Large Language Models (MLLMs) have demonstrated remarkable capabilities in executing instructions for a variety of single-image tasks. Despite this progress, significant challenges remain in modeling long image sequences. In this work, we introduce the versatile multi-modal large language model, mPLUG-Owl3, which enhances the capability for long image-sequence understanding in scenarios that incorporate retrieved image-text knowledge, interleaved image-text, and lengthy videos. Specifically, we propose novel hyper attention blocks to efficiently integrate vision and language into a"},"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":"2408.04840","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-08-09T03:25:42Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"922f15cd8275dc2769ec70c65ef208e7df8e5c3fb9f646e8011b52fcd36bc905","abstract_canon_sha256":"00cbff7b9511a640ac80b697fc68933d455b429e0c46527f4fdfed3e933c988a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T06:14:02.469408Z","signature_b64":"/BLBz/uwbSumJtyEQW40Wp4We21BIaLEjihY9nIas8SPBpkHS115uHSrkigeK8ueEkVMldA9WOAqnjh0ln85BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2515fa79b030e24aa537690ed572d534959f751abd449362a66478b907b77c50","last_reissued_at":"2026-05-20T06:14:02.467716Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T06:14:02.467716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"mPLUG-Owl3: Towards Long Image-Sequence Understanding in Multi-Modal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Anwen Hu, Fei Huang, Haiyang Xu, Haowei Liu, Jiabo Ye, Jingren Zhou, Ji Zhang, Ming Yan, Qi Qian","submitted_at":"2024-08-09T03:25:42Z","abstract_excerpt":"Multi-modal Large Language Models (MLLMs) have demonstrated remarkable capabilities in executing instructions for a variety of single-image tasks. Despite this progress, significant challenges remain in modeling long image sequences. In this work, we introduce the versatile multi-modal large language model, mPLUG-Owl3, which enhances the capability for long image-sequence understanding in scenarios that incorporate retrieved image-text knowledge, interleaved image-text, and lengthy videos. Specifically, we propose novel hyper attention blocks to efficiently integrate vision and language into a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.04840","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/2408.04840/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":"2408.04840","created_at":"2026-05-20T06:14:02.467790+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.04840v2","created_at":"2026-05-20T06:14:02.467790+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.04840","created_at":"2026-05-20T06:14:02.467790+00:00"},{"alias_kind":"pith_short_12","alias_value":"EUK7U6NQGDRE","created_at":"2026-05-20T06:14:02.467790+00:00"},{"alias_kind":"pith_short_16","alias_value":"EUK7U6NQGDREVJJX","created_at":"2026-05-20T06:14:02.467790+00:00"},{"alias_kind":"pith_short_8","alias_value":"EUK7U6NQ","created_at":"2026-05-20T06:14:02.467790+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":25,"internal_anchor_count":25,"sample":[{"citing_arxiv_id":"2411.16771","citing_title":"VidHal: Benchmarking Temporal Hallucinations in Vision LLMs","ref_index":60,"is_internal_anchor":true},{"citing_arxiv_id":"2505.16416","citing_title":"Circle-RoPE: Cone-like Decoupled Rotary Positional Embedding for Large Vision-Language Models","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19846","citing_title":"FineBench: Benchmarking and Enhancing Vision-Language Models for Fine-grained Human Activity Understanding","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18160","citing_title":"Vision Inference Former: Sustaining Visual Consistency in Multimodal Large Language Models","ref_index":44,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19846","citing_title":"FineBench: Benchmarking and Enhancing Vision-Language Models for Fine-grained Human Activity Understanding","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20165","citing_title":"CaMo: Camera Motion Grounded Evaluation and Training for Vision-Language Models","ref_index":78,"is_internal_anchor":true},{"citing_arxiv_id":"2406.08035","citing_title":"LVBench: An Extreme Long Video Understanding Benchmark","ref_index":44,"is_internal_anchor":true},{"citing_arxiv_id":"2506.05425","citing_title":"SIV-Bench: A Video Benchmark for Social Interaction Understanding and Reasoning","ref_index":57,"is_internal_anchor":true},{"citing_arxiv_id":"2506.05831","citing_title":"HeartcareGPT: A Unified Multimodal ECG Suite for Dual Signal-Image Modeling and Understanding","ref_index":5,"is_internal_anchor":true},{"citing_arxiv_id":"2507.00748","citing_title":"Improving the Reasoning of Multi-Image Grounding in MLLMs via Reinforcement Learning","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2509.15602","citing_title":"TennisTV: Do Multimodal Large Language Models Understand Tennis Rallies?","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2510.20093","citing_title":"StableSketcher: Enhancing Diffusion Model for Pixel-based Sketch Generation via Visual Question Answering Feedback","ref_index":36,"is_internal_anchor":true},{"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":143,"is_internal_anchor":true},{"citing_arxiv_id":"2502.04326","citing_title":"WorldSense: Evaluating Real-world Omnimodal Understanding for Multimodal LLMs","ref_index":79,"is_internal_anchor":true},{"citing_arxiv_id":"2501.12386","citing_title":"InternVideo2.5: Empowering Video MLLMs with Long and Rich Context Modeling","ref_index":34,"is_internal_anchor":true},{"citing_arxiv_id":"2603.27259","citing_title":"Seeing the Scene Matters: Revealing Forgetting in Video Understanding Models with a Scene-Aware Long-Video Benchmark","ref_index":56,"is_internal_anchor":true},{"citing_arxiv_id":"2604.22498","citing_title":"CGC: Compositional Grounded Contrast for Fine-Grained Multi-Image Understanding","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05848","citing_title":"VideoRouter: Query-Adaptive Dual Routing for Efficient Long-Video Understanding","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2505.14362","citing_title":"DeepEyes: Incentivizing \"Thinking with Images\" via Reinforcement Learning","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07914","citing_title":"Mitigating Entangled Steering in Large Vision-Language Models for Hallucination Reduction","ref_index":48,"is_internal_anchor":true},{"citing_arxiv_id":"2605.05848","citing_title":"VideoRouter: Query-Adaptive Dual Routing for Efficient Long-Video Understanding","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2604.05418","citing_title":"VideoStir: Understanding Long Videos via Spatio-Temporally Structured and Intent-Aware RAG","ref_index":81,"is_internal_anchor":true},{"citing_arxiv_id":"2604.04379","citing_title":"Reinforce to Learn, Elect to Reason: A Dual Paradigm for Video Reasoning","ref_index":48,"is_internal_anchor":true},{"citing_arxiv_id":"2604.14149","citing_title":"One Token per Highly Selective Frame: Towards Extreme Compression for Long Video Understanding","ref_index":73,"is_internal_anchor":true},{"citing_arxiv_id":"2604.17087","citing_title":"EvoComp: Learning Visual Token Compression for Multimodal Large Language Models via Semantic-Guided Evolutionary Labeling","ref_index":57,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EUK7U6NQGDREVJJXNEHNK4WVGS","json":"https://pith.science/pith/EUK7U6NQGDREVJJXNEHNK4WVGS.json","graph_json":"https://pith.science/api/pith-number/EUK7U6NQGDREVJJXNEHNK4WVGS/graph.json","events_json":"https://pith.science/api/pith-number/EUK7U6NQGDREVJJXNEHNK4WVGS/events.json","paper":"https://pith.science/paper/EUK7U6NQ"},"agent_actions":{"view_html":"https://pith.science/pith/EUK7U6NQGDREVJJXNEHNK4WVGS","download_json":"https://pith.science/pith/EUK7U6NQGDREVJJXNEHNK4WVGS.json","view_paper":"https://pith.science/paper/EUK7U6NQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.04840&json=true","fetch_graph":"https://pith.science/api/pith-number/EUK7U6NQGDREVJJXNEHNK4WVGS/graph.json","fetch_events":"https://pith.science/api/pith-number/EUK7U6NQGDREVJJXNEHNK4WVGS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EUK7U6NQGDREVJJXNEHNK4WVGS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EUK7U6NQGDREVJJXNEHNK4WVGS/action/storage_attestation","attest_author":"https://pith.science/pith/EUK7U6NQGDREVJJXNEHNK4WVGS/action/author_attestation","sign_citation":"https://pith.science/pith/EUK7U6NQGDREVJJXNEHNK4WVGS/action/citation_signature","submit_replication":"https://pith.science/pith/EUK7U6NQGDREVJJXNEHNK4WVGS/action/replication_record"}},"created_at":"2026-05-20T06:14:02.467790+00:00","updated_at":"2026-05-20T06:14:02.467790+00:00"}