Pith Number

pith:LWK7VBFA

pith:2026:LWK7VBFA7CBQDKVKVILWGL6A57

not attested not anchored not stored refs pending

DPRM: A Plug-in Doob h transform-induced Token-Ordering Module for Diffusion Language Models

Andi Han, Atsushi Nitanda, Dake Bu, Hau-San Wong, Qingfu Zhang, Taiji Suzuki, Wei Huang

DPRM introduces a plug-in module that shifts token ordering in diffusion language models from confidence rules to Doob h-transform process reward guidance.

arxiv:2604.24357 v2 · 2026-04-27 · 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{LWK7VBFA7CBQDKVKVILWGL6A57}

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

DPRM improves over confidence-based baselines in pretraining, post-training, test-time scaling, and single-cell masked diffusion, with particularly strong gains on harder reasoning subsets. In protein, molecular generation and DNA design, the effect is more multi-objective: ordering-aware variants significantly improve selected structural or fragment-constrained metrics while not uniformly dominating the host baseline on every quality metric.

C2weakest assumption

That the online bucketized controller tracks the exact DPRM score at empirical-Bernstein rates and that tractable optimization assumptions hold to deliver sample-complexity advantage over random and confidence-only ordering.

C3one line summary

DPRM introduces a Doob h-transform process reward module as a plug-in for token ordering in diffusion language models, with convergence proofs and empirical gains over confidence baselines especially on hard reasoning and scientific design tasks.

Receipt and verification

First computed	2026-06-19T16:09:58.559402Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

5d95fa84a0f88301aaaaaa17632fc0eff9c44a5eb7e213828739691fad4c1246

Aliases

arxiv: 2604.24357 · arxiv_version: 2604.24357v2 · doi: 10.48550/arxiv.2604.24357 · pith_short_12: LWK7VBFA7CBQ · pith_short_16: LWK7VBFA7CBQDKVK · pith_short_8: LWK7VBFA

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/LWK7VBFA7CBQDKVKVILWGL6A57 \
  | 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: 5d95fa84a0f88301aaaaaa17632fc0eff9c44a5eb7e213828739691fad4c1246

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "071a7288cc5839152eda0dfaf391407930a80f73c3b18801f55879ce8b95ff49",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.LG",
    "submitted_at": "2026-04-27T11:50:26Z",
    "title_canon_sha256": "49042bc26768a5363f48d7758d734ea584e7de7c8e22c1743c4ce050639a39ed"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2604.24357",
    "kind": "arxiv",
    "version": 2
  }
}