pith:OMZRZRQA
D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models
Step-distilled diffusion models can learn new concepts through on-policy self-distillation without losing their few-step speed.
arxiv:2605.05204 v2 · 2026-05-06 · cs.CV
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{OMZRZRQAXJWPO3XTE6QET63URH}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
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.
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.
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.
Cited by
Receipt and verification
| First computed | 2026-05-20T00:04:34.372866Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
73331cc600ba6cf76ef327a049fb7489ef586457619f73b28b1b16b7efe1a70d
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/OMZRZRQAXJWPO3XTE6QET63URH \
| jq -c '.canonical_record' \
| python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 73331cc600ba6cf76ef327a049fb7489ef586457619f73b28b1b16b7efe1a70d
Canonical record JSON
{
"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
}
}