{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GISIAEBWSIZUHKDTSHPKMNW523","short_pith_number":"pith:GISIAEBW","schema_version":"1.0","canonical_sha256":"3224801036923343a87391dea636ddd6fc91b8070a6272b267ad1feda8e5b4f6","source":{"kind":"arxiv","id":"2503.06674","version":2},"attestation_state":"computed","paper":{"title":"Learning Few-Step Diffusion Models by Trajectory Distribution Matching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiacheng Sun, Jing Tang, Tianyang Hu, Yihong Luo, Yujun Cai","submitted_at":"2025-03-09T15:53:49Z","abstract_excerpt":"Accelerating diffusion model sampling is crucial for efficient AIGC deployment. While diffusion distillation methods -- based on distribution matching and trajectory matching -- reduce sampling to as few as one step, they fall short on complex tasks like text-to-image generation. Few-step generation offers a better balance between speed and quality, but existing approaches face a persistent trade-off: distribution matching lacks flexibility for multi-step sampling, while trajectory matching often yields suboptimal image quality. To bridge this gap, we propose learning few-step diffusion models"},"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":"2503.06674","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-03-09T15:53:49Z","cross_cats_sorted":[],"title_canon_sha256":"e2b576d608519371df66774d35119e7f5b795f4e5dce40d0e7f693f54ec45bc6","abstract_canon_sha256":"fc623e0fce7508bd5adb6ee8ec5cf900943b80bd7d6b1b536b770b4e365e77c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:29:37.051005Z","signature_b64":"N9kX/LUTolzCPTrMUNQkWW0k3d42cPgtVyfrnhZ0gPBoJKQTvJQ3jGIckj7RNvOjMpE5jZ7nVLW4SFxjeANPCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3224801036923343a87391dea636ddd6fc91b8070a6272b267ad1feda8e5b4f6","last_reissued_at":"2026-07-05T10:29:37.050276Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:29:37.050276Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning Few-Step Diffusion Models by Trajectory Distribution Matching","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiacheng Sun, Jing Tang, Tianyang Hu, Yihong Luo, Yujun Cai","submitted_at":"2025-03-09T15:53:49Z","abstract_excerpt":"Accelerating diffusion model sampling is crucial for efficient AIGC deployment. While diffusion distillation methods -- based on distribution matching and trajectory matching -- reduce sampling to as few as one step, they fall short on complex tasks like text-to-image generation. Few-step generation offers a better balance between speed and quality, but existing approaches face a persistent trade-off: distribution matching lacks flexibility for multi-step sampling, while trajectory matching often yields suboptimal image quality. To bridge this gap, we propose learning few-step diffusion models"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.06674","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/2503.06674/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":"2503.06674","created_at":"2026-07-05T10:29:37.050382+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.06674v2","created_at":"2026-07-05T10:29:37.050382+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.06674","created_at":"2026-07-05T10:29:37.050382+00:00"},{"alias_kind":"pith_short_12","alias_value":"GISIAEBWSIZU","created_at":"2026-07-05T10:29:37.050382+00:00"},{"alias_kind":"pith_short_16","alias_value":"GISIAEBWSIZUHKDT","created_at":"2026-07-05T10:29:37.050382+00:00"},{"alias_kind":"pith_short_8","alias_value":"GISIAEBW","created_at":"2026-07-05T10:29:37.050382+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.05204","citing_title":"D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2602.03139","citing_title":"Diversity-Preserved Distribution Matching Distillation for Fast Visual Synthesis","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05204","citing_title":"D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17834","citing_title":"Stabilizing, Scaling & Enhancing MeanFlow for Large-scale Diffusion Distillation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2512.04677","citing_title":"Live Avatar: Streaming Real-time Audio-Driven Avatar Generation with Infinite Length","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2510.02283","citing_title":"Self-Forcing++: Towards Minute-Scale High-Quality Video Generation","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2512.16776","citing_title":"Kling-Omni Technical Report","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04018","citing_title":"1.x-Distill: Breaking the Diversity, Quality, and Efficiency Barrier in Distribution Matching Distillation","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23632","citing_title":"Hallo-Live: Real-Time Streaming Joint Audio-Video Avatar Generation with Asynchronous Dual-Stream and Human-Centric Preference Distillation","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.05204","citing_title":"D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14580","citing_title":"TurboTalk: Progressive Distillation for One-Step Audio-Driven Talking Avatar Generation","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15911","citing_title":"Efficient Video Diffusion Models: Advancements and Challenges","ref_index":97,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GISIAEBWSIZUHKDTSHPKMNW523","json":"https://pith.science/pith/GISIAEBWSIZUHKDTSHPKMNW523.json","graph_json":"https://pith.science/api/pith-number/GISIAEBWSIZUHKDTSHPKMNW523/graph.json","events_json":"https://pith.science/api/pith-number/GISIAEBWSIZUHKDTSHPKMNW523/events.json","paper":"https://pith.science/paper/GISIAEBW"},"agent_actions":{"view_html":"https://pith.science/pith/GISIAEBWSIZUHKDTSHPKMNW523","download_json":"https://pith.science/pith/GISIAEBWSIZUHKDTSHPKMNW523.json","view_paper":"https://pith.science/paper/GISIAEBW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.06674&json=true","fetch_graph":"https://pith.science/api/pith-number/GISIAEBWSIZUHKDTSHPKMNW523/graph.json","fetch_events":"https://pith.science/api/pith-number/GISIAEBWSIZUHKDTSHPKMNW523/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GISIAEBWSIZUHKDTSHPKMNW523/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GISIAEBWSIZUHKDTSHPKMNW523/action/storage_attestation","attest_author":"https://pith.science/pith/GISIAEBWSIZUHKDTSHPKMNW523/action/author_attestation","sign_citation":"https://pith.science/pith/GISIAEBWSIZUHKDTSHPKMNW523/action/citation_signature","submit_replication":"https://pith.science/pith/GISIAEBWSIZUHKDTSHPKMNW523/action/replication_record"}},"created_at":"2026-07-05T10:29:37.050382+00:00","updated_at":"2026-07-05T10:29:37.050382+00:00"}