{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZG6OA227PFP4OFYASGTHCAK2NJ","short_pith_number":"pith:ZG6OA227","schema_version":"1.0","canonical_sha256":"c9bce06b5f795fc7170091a671015a6a5718aa86aa85f8b65443e3d7c333e12a","source":{"kind":"arxiv","id":"2506.23044","version":2},"attestation_state":"computed","paper":{"title":"Ovis-U1 Technical Report","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Guo-Hua Wang, Jianshan Zhao, Liangfu Cao, Lunhao Duan, Minghao Fu, Pengxin Zhan, Qing-Guo Chen, Shanshan Zhao, Shiyin Lu, Xiaohao Chen, Xinjie Zhang, Yang Li","submitted_at":"2025-06-29T00:40:17Z","abstract_excerpt":"In this report, we introduce Ovis-U1, a 3-billion-parameter unified model that integrates multimodal understanding, text-to-image generation, and image editing capabilities. Building on the foundation of the Ovis series, Ovis-U1 incorporates a diffusion-based visual decoder paired with a bidirectional token refiner, enabling image generation tasks comparable to leading models like GPT-4o. Unlike some previous models that use a frozen MLLM for generation tasks, Ovis-U1 utilizes a new unified training approach starting from a language model. Compared to training solely on understanding or genera"},"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":"2506.23044","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-29T00:40:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5debf616c9caa1843af74a85d6f10cfbaefe4556b734a5ed0197be6224459a6e","abstract_canon_sha256":"1bb743647be33becd935d3a9823e90f6a06e48d48218ed248632d7aed3ec9acd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:30:03.654998Z","signature_b64":"1DtpTfpOPEyRubxG9a3xEcw2ty3khppYonHwlrX+CcG3bQHh8EnKzXbBb9/C+a5h8QTBctVa6M//NuVRMyqgAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c9bce06b5f795fc7170091a671015a6a5718aa86aa85f8b65443e3d7c333e12a","last_reissued_at":"2026-07-05T11:30:03.654499Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:30:03.654499Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Ovis-U1 Technical Report","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Guo-Hua Wang, Jianshan Zhao, Liangfu Cao, Lunhao Duan, Minghao Fu, Pengxin Zhan, Qing-Guo Chen, Shanshan Zhao, Shiyin Lu, Xiaohao Chen, Xinjie Zhang, Yang Li","submitted_at":"2025-06-29T00:40:17Z","abstract_excerpt":"In this report, we introduce Ovis-U1, a 3-billion-parameter unified model that integrates multimodal understanding, text-to-image generation, and image editing capabilities. Building on the foundation of the Ovis series, Ovis-U1 incorporates a diffusion-based visual decoder paired with a bidirectional token refiner, enabling image generation tasks comparable to leading models like GPT-4o. Unlike some previous models that use a frozen MLLM for generation tasks, Ovis-U1 utilizes a new unified training approach starting from a language model. Compared to training solely on understanding or genera"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.23044","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/2506.23044/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":"2506.23044","created_at":"2026-07-05T11:30:03.654558+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.23044v2","created_at":"2026-07-05T11:30:03.654558+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.23044","created_at":"2026-07-05T11:30:03.654558+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZG6OA227PFP4","created_at":"2026-07-05T11:30:03.654558+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZG6OA227PFP4OFYA","created_at":"2026-07-05T11:30:03.654558+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZG6OA227","created_at":"2026-07-05T11:30:03.654558+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":18,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.17565","citing_title":"UniGeo: Unifying Geometric Guidance for Camera-Controllable Image Editing via Video Models","ref_index":68,"is_internal_anchor":true},{"citing_arxiv_id":"2606.19531","citing_title":"ImageWAM: Do World Action Models Really Need Video Generation, or Just Image Editing?","ref_index":86,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13679","citing_title":"InterleaveThinker: Reinforcing Agentic Interleaved Generation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01022","citing_title":"ProductWebGen: Benchmarking Multimodal Product Webpage Generation","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18678","citing_title":"Lance: Unified Multimodal Modeling by Multi-Task Synergy","ref_index":113,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24763","citing_title":"Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18678","citing_title":"Lance: Unified Multimodal Modeling by Multi-Task Synergy","ref_index":112,"is_internal_anchor":false},{"citing_arxiv_id":"2602.01554","citing_title":"InfoTok: Information-Theoretic Regularization for Capacity-Constrained Shared Visual Tokenization in Unified MLLMs","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2602.07064","citing_title":"OmniFysics: Towards Physical Intelligence Evolution via Omni-Modal Signal Processing and Network Optimization","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2602.06663","citing_title":"PlanViz: Evaluating Planning-Oriented Image Generation and Editing for Computer-Use Tasks","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12500","citing_title":"SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-unify Architecture","ref_index":130,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11400","citing_title":"UniPath: Adaptive Coordination of Understanding and Generation for Unified Multimodal Reasoning","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17565","citing_title":"UniGeo: Unifying Geometric Guidance for Camera-Controllable Image Editing via Video Models","ref_index":68,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24763","citing_title":"Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10949","citing_title":"Pseudo-Unification: Entropy Probing Reveals Divergent Information Patterns in Unified Multimodal Models","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04487","citing_title":"Training-Free Image Editing with Visual Context Integration and Concept Alignment","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17850","citing_title":"UniCSG: Unified High-Fidelity Content-Constrained Style-Driven Generation via Staged Semantic and Frequency Disentanglement","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17565","citing_title":"UniGeo: Unifying Geometric Guidance for Camera-Controllable Image Editing via Video Models","ref_index":68,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZG6OA227PFP4OFYASGTHCAK2NJ","json":"https://pith.science/pith/ZG6OA227PFP4OFYASGTHCAK2NJ.json","graph_json":"https://pith.science/api/pith-number/ZG6OA227PFP4OFYASGTHCAK2NJ/graph.json","events_json":"https://pith.science/api/pith-number/ZG6OA227PFP4OFYASGTHCAK2NJ/events.json","paper":"https://pith.science/paper/ZG6OA227"},"agent_actions":{"view_html":"https://pith.science/pith/ZG6OA227PFP4OFYASGTHCAK2NJ","download_json":"https://pith.science/pith/ZG6OA227PFP4OFYASGTHCAK2NJ.json","view_paper":"https://pith.science/paper/ZG6OA227","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.23044&json=true","fetch_graph":"https://pith.science/api/pith-number/ZG6OA227PFP4OFYASGTHCAK2NJ/graph.json","fetch_events":"https://pith.science/api/pith-number/ZG6OA227PFP4OFYASGTHCAK2NJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZG6OA227PFP4OFYASGTHCAK2NJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZG6OA227PFP4OFYASGTHCAK2NJ/action/storage_attestation","attest_author":"https://pith.science/pith/ZG6OA227PFP4OFYASGTHCAK2NJ/action/author_attestation","sign_citation":"https://pith.science/pith/ZG6OA227PFP4OFYASGTHCAK2NJ/action/citation_signature","submit_replication":"https://pith.science/pith/ZG6OA227PFP4OFYASGTHCAK2NJ/action/replication_record"}},"created_at":"2026-07-05T11:30:03.654558+00:00","updated_at":"2026-07-05T11:30:03.654558+00:00"}