Pith Number

pith:SEMDPEKU

pith:2026:SEMDPEKUE5WZB6HAAWJMOI22AD

not attested not anchored not stored refs pending

Simple Self-Conditioning Adaptation for Masked Diffusion Models

Ferdinando Fioretto, Huu Binh Ta, Michael Cardei

A post-training adaptation conditions masked diffusion models on their own prior clean predictions to enable repeated refinement across denoising steps.

arxiv:2604.26985 v2 · 2026-04-28 · cs.LG · cs.AI

Open paper page JSON Open Graph Bundle Merged state Verified badge What is a Pith Number?

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{SEMDPEKUE5WZB6HAAWJMOI22AD}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge

Record completeness

1 Bitcoin timestamp

2 Internet Archive

3 Author claim open · sign in to claim

4 Citations open

5 Replications open

✓ Portable graph bundle live · download bundle · merged state

The bundle contains the canonical record plus signed events. A mirror can host it anywhere and recompute the same current state with the deterministic merge algorithm.

Claims

C1strongest claim

SCMDM achieves nearly a 50% reduction in generative perplexity on OWT-trained models (42.89 to 23.72), alongside strong improvements in discretized image synthesis quality, small molecular generation, and enhanced fidelity in genomic distribution modeling.

C2weakest assumption

Once the model's self-generated clean-state estimates become informative, specialization to refinement is preferable to mixing conditional and unconditional objectives in the post-training regime.

C3one line summary

SCMDM adapts trained masked diffusion models to condition denoising steps on their own prior clean predictions, cutting generative perplexity nearly in half on open-web text while improving discretized image, molecule, and genomic synthesis.

Cited by

1 paper in Pith

Low Perplexity is Repetition: A One-Dimensional Self-Conditioning Attractor in Continuous Diffusion LMs

Receipt and verification

First computed	2026-06-09T01:05:18.389213Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

9118379154276d90f8e00592c7235a00ee279d68d87bf4eb609a3b1a3e1e44aa

Aliases

arxiv: 2604.26985 · arxiv_version: 2604.26985v2 · doi: 10.48550/arxiv.2604.26985 · pith_short_12: SEMDPEKUE5WZ · pith_short_16: SEMDPEKUE5WZB6HA · pith_short_8: SEMDPEKU

Agent API

Resolver JSON Graph JSON Events JSON Schema Signing key

Verify this Pith Number yourself

curl -sH 'Accept: application/ld+json' https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD \
  | 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: 9118379154276d90f8e00592c7235a00ee279d68d87bf4eb609a3b1a3e1e44aa

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "1e0a09f4cc10b8132d974ba9fb7b554b49bf4fed6f7320b3f480af994e0181aa",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-04-28T19:34:04Z",
    "title_canon_sha256": "f25407a5fbb1d96396163dc29404036f6006b0d379e78e3e1e6beb4ed054a58b"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.26985",
    "kind": "arxiv",
    "version": 2
  }
}