Pith Number

pith:Y6GGGRXY

pith:2025:Y6GGGRXYR2JWCVBXGJGF3UXU6P

not attested not anchored not stored refs pending

CreditDecoding: Accelerating Parallel Decoding in Diffusion Large Language Models with Trace Credit

Haibo Feng, Jianguo Li, Kangyu Wang, Lin Liu, Weijia Zhao, Weiyao Lin, Zhenzhong Lan, Zhiyun Jiang

By accumulating historical prediction evidence, CreditDecoding lets diffusion LLMs accept correct tokens earlier in parallel decoding, delivering up to 5.48 times speedup and modest accuracy gains.

arxiv:2510.06133 v3 · 2025-10-07 · cs.CL · 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{Y6GGGRXYR2JWCVBXGJGF3UXU6P}

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

On eight benchmarks, CreditDecoding achieves up to 5.48 times speedup with +0.48 accuracy on LLaDA-8B and consistently improves performance across diverse dLLM architectures and parameter scales.

C2weakest assumption

The assumption that accumulating historical prediction evidence via Trace Credit and fusing it with current logits will reliably increase acceptance of correct tokens without introducing new errors or degrading performance on positions that were previously handled correctly.

C3one line summary

CreditDecoding accelerates parallel decoding in diffusion LLMs by fusing accumulated Trace Credit with current logits to accept early-correct tokens sooner, yielding up to 5.48x speedup and accuracy gains.

Formal links

2 machine-checked theorem links

Cited by

4 papers in Pith

DMax: Aggressive Parallel Decoding for dLLMs

TAD: Temporal-Aware Trajectory Self-Distillation for Fast and Accurate Diffusion LLM

$R^2$-dLLM: Accelerating Diffusion Large Language Models via Spatio-Temporal Redundancy Reduction

DMax: Aggressive Parallel Decoding for dLLMs

Receipt and verification

First computed	2026-05-27T01:04:52.061714Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

c78c6346f88e93615437324c5dd2f4f3e060e4192f8946155b3f610f2a485f5d

Aliases

arxiv: 2510.06133 · arxiv_version: 2510.06133v3 · doi: 10.48550/arxiv.2510.06133 · pith_short_12: Y6GGGRXYR2JW · pith_short_16: Y6GGGRXYR2JWCVBX · pith_short_8: Y6GGGRXY

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/Y6GGGRXYR2JWCVBXGJGF3UXU6P \
  | 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: c78c6346f88e93615437324c5dd2f4f3e060e4192f8946155b3f610f2a485f5d

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "145a05375e762d97428eae8c402d60eb90b0c7848e32b5c915fc966b41da9894",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2025-10-07T17:08:33Z",
    "title_canon_sha256": "d21668d3b0d1a1456f303fd179ba4b3da5437d63dcd2874a8db986eef1d07fe3"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2510.06133",
    "kind": "arxiv",
    "version": 3
  }
}