{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:OMZRZRQAXJWPO3XTE6QET63URH","short_pith_number":"pith:OMZRZRQA","canonical_record":{"source":{"id":"2605.05204","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-06T17:59:34Z","cross_cats_sorted":[],"title_canon_sha256":"228087488510f3779d05703507fcbca4d9c3923489b87ae946c343d8183b7cd9","abstract_canon_sha256":"7aff7bbedc1e8517050125d9d897db7c966ef91866f92ef413ee7747ee0fd44b"},"schema_version":"1.0"},"canonical_sha256":"73331cc600ba6cf76ef327a049fb7489ef586457619f73b28b1b16b7efe1a70d","source":{"kind":"arxiv","id":"2605.05204","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.05204","created_at":"2026-05-20T00:04:34Z"},{"alias_kind":"arxiv_version","alias_value":"2605.05204v2","created_at":"2026-05-20T00:04:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.05204","created_at":"2026-05-20T00:04:34Z"},{"alias_kind":"pith_short_12","alias_value":"OMZRZRQAXJWP","created_at":"2026-05-20T00:04:34Z"},{"alias_kind":"pith_short_16","alias_value":"OMZRZRQAXJWPO3XT","created_at":"2026-05-20T00:04:34Z"},{"alias_kind":"pith_short_8","alias_value":"OMZRZRQA","created_at":"2026-05-20T00:04:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:OMZRZRQAXJWPO3XTE6QET63URH","target":"record","payload":{"canonical_record":{"source":{"id":"2605.05204","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-06T17:59:34Z","cross_cats_sorted":[],"title_canon_sha256":"228087488510f3779d05703507fcbca4d9c3923489b87ae946c343d8183b7cd9","abstract_canon_sha256":"7aff7bbedc1e8517050125d9d897db7c966ef91866f92ef413ee7747ee0fd44b"},"schema_version":"1.0"},"canonical_sha256":"73331cc600ba6cf76ef327a049fb7489ef586457619f73b28b1b16b7efe1a70d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:04:34.373609Z","signature_b64":"i1EiSN1cQSm+XPHNxXv63Ji7rP0B8LFVCa8G9uuMgmbSGT5GHiJE7i1nAFJ6fK0LbE8YyG0V+l9eeB2CQU2DAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"73331cc600ba6cf76ef327a049fb7489ef586457619f73b28b1b16b7efe1a70d","last_reissued_at":"2026-05-20T00:04:34.372866Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:04:34.372866Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.05204","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JAGYxjKOmdmfjxonhixjnNB5XMHpRM9e7g07cWTFicJ67DBIMo+qXLIuzG9OVTjjWe3ZBS3hyRGU2lcHFDTQDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T10:02:18.514949Z"},"content_sha256":"c1a3d1e69139c4ae9ac91502223b97b3e3dae5a55bfc024d98cc9f331e06f50b","schema_version":"1.0","event_id":"sha256:c1a3d1e69139c4ae9ac91502223b97b3e3dae5a55bfc024d98cc9f331e06f50b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:OMZRZRQAXJWPO3XTE6QET63URH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Step-distilled diffusion models can learn new concepts through on-policy self-distillation without losing their few-step speed.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dengyang Jiang, Dongyang Liu, Harry Yang, Mingzhe Zheng, Peng Gao, Qilong Wu, Ruoyi Du, Steven Hoi, Xiangpeng Yang, Xin Jin, Zanyi Wang, Zhen Li","submitted_at":"2026-05-06T17:59:34Z","abstract_excerpt":"The landscape of high-performance image generation models is currently shifting from the inefficient multi-step ones to the efficient few-step counterparts (e.g, Z-Image-Turbo and FLUX.2-klein). However, these models present significant challenges for direct continuous supervised fine-tuning. For example, applying the commonly used fine-tuning technique would compromise their inherent few-step inference capability. To address this, we propose D-OPSD, a novel training paradigm for step-distilled diffusion models that enables on-policy learning during supervised fine-tuning. We first find that t"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"By optimized on the model's own trajectory and under it's own supervision, D-OPSD enables the model to learn new concept, style, etc. without sacrificing the original few-step capacity.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The modern diffusion model where the LLM/VLM serves as the encoder can inherit its encoder's in-context capabilities. This is stated as the key finding that enables treating training as an on-policy self-distillation process.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"D-OPSD enables continuous supervised fine-tuning of few-step diffusion models via on-policy self-distillation where the model acts as both teacher (multimodal context) and student (text-only context) on its own roll-outs.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Step-distilled diffusion models can learn new concepts through on-policy self-distillation without losing their few-step speed.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"0b5f630debbacace2b6c1158d075e455381f0a0c345d6240a8ae11c6e35fc2ee"},"source":{"id":"2605.05204","kind":"arxiv","version":2},"verdict":{"id":"e67d005f-0d90-42e6-866b-839b34035cc1","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T17:21:25.679765Z","strongest_claim":"By optimized on the model's own trajectory and under it's own supervision, D-OPSD enables the model to learn new concept, style, etc. without sacrificing the original few-step capacity.","one_line_summary":"D-OPSD enables continuous supervised fine-tuning of few-step diffusion models via on-policy self-distillation where the model acts as both teacher (multimodal context) and student (text-only context) on its own roll-outs.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The modern diffusion model where the LLM/VLM serves as the encoder can inherit its encoder's in-context capabilities. This is stated as the key finding that enables treating training as an on-policy self-distillation process.","pith_extraction_headline":"Step-distilled diffusion models can learn new concepts through on-policy self-distillation without losing their few-step speed."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.05204/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_title_agreement","ran_at":"2026-05-19T21:01:19.701997Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T13:46:29.903248Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"7a93af0ac77f5e5d70472920ba84c19870df33200af4ca95d6da1186d6da58dd"},"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"},"verdict_id":"e67d005f-0d90-42e6-866b-839b34035cc1"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:04:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dFrlyY6BFrODb6kunSIV6NRVGoo6gCgAWWx3NO8NkiKTr4DaQWfDo0vWnXZNDtmZqZ2MXiTOF9sGHGS3QdteBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T10:02:18.515947Z"},"content_sha256":"bf143445b81a9db2eb59daeebfec120594fd232b5e859707a361cd4c01c43321","schema_version":"1.0","event_id":"sha256:bf143445b81a9db2eb59daeebfec120594fd232b5e859707a361cd4c01c43321"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OMZRZRQAXJWPO3XTE6QET63URH/bundle.json","state_url":"https://pith.science/pith/OMZRZRQAXJWPO3XTE6QET63URH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OMZRZRQAXJWPO3XTE6QET63URH/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-27T10:02:18Z","links":{"resolver":"https://pith.science/pith/OMZRZRQAXJWPO3XTE6QET63URH","bundle":"https://pith.science/pith/OMZRZRQAXJWPO3XTE6QET63URH/bundle.json","state":"https://pith.science/pith/OMZRZRQAXJWPO3XTE6QET63URH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OMZRZRQAXJWPO3XTE6QET63URH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:OMZRZRQAXJWPO3XTE6QET63URH","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"7aff7bbedc1e8517050125d9d897db7c966ef91866f92ef413ee7747ee0fd44b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-06T17:59:34Z","title_canon_sha256":"228087488510f3779d05703507fcbca4d9c3923489b87ae946c343d8183b7cd9"},"schema_version":"1.0","source":{"id":"2605.05204","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.05204","created_at":"2026-05-20T00:04:34Z"},{"alias_kind":"arxiv_version","alias_value":"2605.05204v2","created_at":"2026-05-20T00:04:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.05204","created_at":"2026-05-20T00:04:34Z"},{"alias_kind":"pith_short_12","alias_value":"OMZRZRQAXJWP","created_at":"2026-05-20T00:04:34Z"},{"alias_kind":"pith_short_16","alias_value":"OMZRZRQAXJWPO3XT","created_at":"2026-05-20T00:04:34Z"},{"alias_kind":"pith_short_8","alias_value":"OMZRZRQA","created_at":"2026-05-20T00:04:34Z"}],"graph_snapshots":[{"event_id":"sha256:bf143445b81a9db2eb59daeebfec120594fd232b5e859707a361cd4c01c43321","target":"graph","created_at":"2026-05-20T00:04:34Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"By optimized on the model's own trajectory and under it's own supervision, D-OPSD enables the model to learn new concept, style, etc. without sacrificing the original few-step capacity."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The modern diffusion model where the LLM/VLM serves as the encoder can inherit its encoder's in-context capabilities. This is stated as the key finding that enables treating training as an on-policy self-distillation process."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"D-OPSD enables continuous supervised fine-tuning of few-step diffusion models via on-policy self-distillation where the model acts as both teacher (multimodal context) and student (text-only context) on its own roll-outs."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Step-distilled diffusion models can learn new concepts through on-policy self-distillation without losing their few-step speed."}],"snapshot_sha256":"0b5f630debbacace2b6c1158d075e455381f0a0c345d6240a8ae11c6e35fc2ee"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T21:01:19.701997Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T13:46:29.903248Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.05204/integrity.json","findings":[],"snapshot_sha256":"7a93af0ac77f5e5d70472920ba84c19870df33200af4ca95d6da1186d6da58dd","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The landscape of high-performance image generation models is currently shifting from the inefficient multi-step ones to the efficient few-step counterparts (e.g, Z-Image-Turbo and FLUX.2-klein). However, these models present significant challenges for direct continuous supervised fine-tuning. For example, applying the commonly used fine-tuning technique would compromise their inherent few-step inference capability. To address this, we propose D-OPSD, a novel training paradigm for step-distilled diffusion models that enables on-policy learning during supervised fine-tuning. We first find that t","authors_text":"Dengyang Jiang, Dongyang Liu, Harry Yang, Mingzhe Zheng, Peng Gao, Qilong Wu, Ruoyi Du, Steven Hoi, Xiangpeng Yang, Xin Jin, Zanyi Wang, Zhen Li","cross_cats":[],"headline":"Step-distilled diffusion models can learn new concepts through on-policy self-distillation without losing their few-step speed.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-06T17:59:34Z","title":"D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.05204","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T17:21:25.679765Z","id":"e67d005f-0d90-42e6-866b-839b34035cc1","model_set":{"reader":"grok-4.3"},"one_line_summary":"D-OPSD enables continuous supervised fine-tuning of few-step diffusion models via on-policy self-distillation where the model acts as both teacher (multimodal context) and student (text-only context) on its own roll-outs.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Step-distilled diffusion models can learn new concepts through on-policy self-distillation without losing their few-step speed.","strongest_claim":"By optimized on the model's own trajectory and under it's own supervision, D-OPSD enables the model to learn new concept, style, etc. without sacrificing the original few-step capacity.","weakest_assumption":"The modern diffusion model where the LLM/VLM serves as the encoder can inherit its encoder's in-context capabilities. This is stated as the key finding that enables treating training as an on-policy self-distillation process."}},"verdict_id":"e67d005f-0d90-42e6-866b-839b34035cc1"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:c1a3d1e69139c4ae9ac91502223b97b3e3dae5a55bfc024d98cc9f331e06f50b","target":"record","created_at":"2026-05-20T00:04:34Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"7aff7bbedc1e8517050125d9d897db7c966ef91866f92ef413ee7747ee0fd44b","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-06T17:59:34Z","title_canon_sha256":"228087488510f3779d05703507fcbca4d9c3923489b87ae946c343d8183b7cd9"},"schema_version":"1.0","source":{"id":"2605.05204","kind":"arxiv","version":2}},"canonical_sha256":"73331cc600ba6cf76ef327a049fb7489ef586457619f73b28b1b16b7efe1a70d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"73331cc600ba6cf76ef327a049fb7489ef586457619f73b28b1b16b7efe1a70d","first_computed_at":"2026-05-20T00:04:34.372866Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:04:34.372866Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"i1EiSN1cQSm+XPHNxXv63Ji7rP0B8LFVCa8G9uuMgmbSGT5GHiJE7i1nAFJ6fK0LbE8YyG0V+l9eeB2CQU2DAQ==","signature_status":"signed_v1","signed_at":"2026-05-20T00:04:34.373609Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.05204","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c1a3d1e69139c4ae9ac91502223b97b3e3dae5a55bfc024d98cc9f331e06f50b","sha256:bf143445b81a9db2eb59daeebfec120594fd232b5e859707a361cd4c01c43321"],"state_sha256":"a04c48af030a31b725e1f861b78798c004002b9c476b66ff3d8316c207733b3e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bw5s9Yw9muP7VwJGpAqMm/czmI4X4dha6l3twWLVAdiWUap+RWNGrKhQnOwvx+zRfHxjgdVWOt5/Ffsi0EC3Bw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T10:02:18.520511Z","bundle_sha256":"29fda3a3a0844342355f18f9d6ba739132e412c6b0fe497e0bfebb90c9a40252"}}