{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:G4P7ADHRAEZ6D6VFSWA3VQJF2Q","short_pith_number":"pith:G4P7ADHR","schema_version":"1.0","canonical_sha256":"371ff00cf10133e1faa59581bac125d438abf2fae83d15d90efaa7804da31d44","source":{"kind":"arxiv","id":"2505.20147","version":3},"attestation_state":"computed","paper":{"title":"FUDOKI: Discrete Flow-based Unified Understanding and Generation via Kinetic-Optimal Velocities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aoxue Li, Chengyue Wu, Jiacheng Sun, Jin Wang, Ning Kang, Ping Luo, Shifeng Zhang, Yao Lai, Zhenguo Li","submitted_at":"2025-05-26T15:46:53Z","abstract_excerpt":"The rapid progress of large language models (LLMs) has catalyzed the emergence of multimodal large language models (MLLMs) that unify visual understanding and image generation within a single framework. However, most existing MLLMs rely on autoregressive (AR) architectures, which impose inherent limitations on future development, such as the raster-scan order in image generation and restricted reasoning abilities in causal context modeling. In this work, we challenge the dominance of AR-based approaches by introducing FUDOKI, a unified multimodal model purely based on discrete flow matching, 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":"2505.20147","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-26T15:46:53Z","cross_cats_sorted":[],"title_canon_sha256":"41ed14626015c2704751af0191b77ba06c8f724fab8fb7c56e9a4cf97674338a","abstract_canon_sha256":"72b168e328b48ad9e547269985ba58e15521f650e24ed49dc7e9f1313ebe33fc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:29.759000Z","signature_b64":"OqrUJlLKD/1Q3hfl3Xro6QX4is3HjWUbrqqtaZCy7zBsn0mffED1yBWXjtdnYTMrcKJb4Uhs2gh3VEMjhxdsAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"371ff00cf10133e1faa59581bac125d438abf2fae83d15d90efaa7804da31d44","last_reissued_at":"2026-07-05T11:42:29.758500Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:29.758500Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FUDOKI: Discrete Flow-based Unified Understanding and Generation via Kinetic-Optimal Velocities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Aoxue Li, Chengyue Wu, Jiacheng Sun, Jin Wang, Ning Kang, Ping Luo, Shifeng Zhang, Yao Lai, Zhenguo Li","submitted_at":"2025-05-26T15:46:53Z","abstract_excerpt":"The rapid progress of large language models (LLMs) has catalyzed the emergence of multimodal large language models (MLLMs) that unify visual understanding and image generation within a single framework. However, most existing MLLMs rely on autoregressive (AR) architectures, which impose inherent limitations on future development, such as the raster-scan order in image generation and restricted reasoning abilities in causal context modeling. In this work, we challenge the dominance of AR-based approaches by introducing FUDOKI, a unified multimodal model purely based on discrete flow matching, a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.20147","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/2505.20147/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":"2505.20147","created_at":"2026-07-05T11:42:29.758562+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.20147v3","created_at":"2026-07-05T11:42:29.758562+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.20147","created_at":"2026-07-05T11:42:29.758562+00:00"},{"alias_kind":"pith_short_12","alias_value":"G4P7ADHRAEZ6","created_at":"2026-07-05T11:42:29.758562+00:00"},{"alias_kind":"pith_short_16","alias_value":"G4P7ADHRAEZ6D6VF","created_at":"2026-07-05T11:42:29.758562+00:00"},{"alias_kind":"pith_short_8","alias_value":"G4P7ADHR","created_at":"2026-07-05T11:42:29.758562+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.18518","citing_title":"UDM-GRPO: Stable and Efficient Group Relative Policy Optimization for Uniform Discrete Diffusion Models","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2606.24333","citing_title":"UniTranslator: A Unified Multi-modal Framework for End-to-end In-Image Machine Translation","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2606.04264","citing_title":"UniCanvas: A Diffusion-base Unified Model for Text-in-Image Joint Generation","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2509.21912","citing_title":"Discrete Guidance Matching: Exact Guidance for Discrete Flow Matching","ref_index":78,"is_internal_anchor":false},{"citing_arxiv_id":"2603.26320","citing_title":"DFM-VLA: Iterative Action Refinement for Robot Manipulation via Discrete Flow Matching","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08882","citing_title":"Discrete Flow Matching: Convergence Guarantees Under Minimal Assumptions","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08724","citing_title":"SynerMedGen: Synergizing Medical Multimodal Understanding with Generation via Task Alignment","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16514","citing_title":"BARD: Bridging AutoRegressive and Diffusion Vision-Language Models Via Highly Efficient Progressive Block Merging and Stage-Wise Distillation","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18518","citing_title":"UDM-GRPO: Stable and Efficient Group Relative Policy Optimization for Uniform Discrete Diffusion Models","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G4P7ADHRAEZ6D6VFSWA3VQJF2Q","json":"https://pith.science/pith/G4P7ADHRAEZ6D6VFSWA3VQJF2Q.json","graph_json":"https://pith.science/api/pith-number/G4P7ADHRAEZ6D6VFSWA3VQJF2Q/graph.json","events_json":"https://pith.science/api/pith-number/G4P7ADHRAEZ6D6VFSWA3VQJF2Q/events.json","paper":"https://pith.science/paper/G4P7ADHR"},"agent_actions":{"view_html":"https://pith.science/pith/G4P7ADHRAEZ6D6VFSWA3VQJF2Q","download_json":"https://pith.science/pith/G4P7ADHRAEZ6D6VFSWA3VQJF2Q.json","view_paper":"https://pith.science/paper/G4P7ADHR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.20147&json=true","fetch_graph":"https://pith.science/api/pith-number/G4P7ADHRAEZ6D6VFSWA3VQJF2Q/graph.json","fetch_events":"https://pith.science/api/pith-number/G4P7ADHRAEZ6D6VFSWA3VQJF2Q/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G4P7ADHRAEZ6D6VFSWA3VQJF2Q/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G4P7ADHRAEZ6D6VFSWA3VQJF2Q/action/storage_attestation","attest_author":"https://pith.science/pith/G4P7ADHRAEZ6D6VFSWA3VQJF2Q/action/author_attestation","sign_citation":"https://pith.science/pith/G4P7ADHRAEZ6D6VFSWA3VQJF2Q/action/citation_signature","submit_replication":"https://pith.science/pith/G4P7ADHRAEZ6D6VFSWA3VQJF2Q/action/replication_record"}},"created_at":"2026-07-05T11:42:29.758562+00:00","updated_at":"2026-07-05T11:42:29.758562+00:00"}