Pith Number

pith:F4JIFFKU

pith:2026:F4JIFFKUB327TU2XITLUJGKXLE

not attested not anchored not stored refs resolved

Differences in Text Generated by Diffusion and Autoregressive Language Models

Chengwei Liang, Jingzhao Zhang, Meiqi Gu, Minrui Luo, Tianxing He, Xingyan Chen, Zeyang Zhang

Diffusion language models generate text with higher semantic coherence and diversity than autoregressive models due to bidirectional context in training, while lower entropy stems from their decoding algorithms.

arxiv:2605.12522 v1 · 2026-04-04 · 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{F4JIFFKUB327TU2XITLUJGKXLE}

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

Results suggest that the DLM training objective contributes to the increases in semantic coherence and semantic diversity, but has a minor influence on entropy. These differences are primarily driven by the bidirectional context; the reduction in entropy stems from DLMs' decoding algorithms, particularly confidence-based remasking strategies.

C2weakest assumption

That the controlled experiments can cleanly decouple training-objective effects from decoding-algorithm effects without confounding factors from implementation choices or data selection.

C3one line summary

DLMs exhibit lower n-gram entropy, higher semantic coherence, and higher semantic diversity than ARMs, primarily due to bidirectional context and remasking decoding strategies.

References

41 extracted · 41 resolved · 18 Pith anchors

[1] Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs · arXiv:2503.01743

[2] Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models · arXiv:2503.09573

[3] Accelerated sampling from masked diffusion models via entropy bounded unmasking.arXiv preprint arXiv:2505.24857

[4] LLaDA2.0: Scaling Up Diffusion Language Models to 100B · arXiv:2512.15745

[5] M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillation · arXiv:2402.03216

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-18T03:10:02.854950Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

2f128295540ef5f9d35744d7449957592839c7d431bff461bb36a16f9aa02b56

Aliases

arxiv: 2605.12522 · arxiv_version: 2605.12522v1 · doi: 10.48550/arxiv.2605.12522 · pith_short_12: F4JIFFKUB327 · pith_short_16: F4JIFFKUB327TU2X · pith_short_8: F4JIFFKU

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/F4JIFFKUB327TU2XITLUJGKXLE \
  | 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: 2f128295540ef5f9d35744d7449957592839c7d431bff461bb36a16f9aa02b56

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "6aa56592c21a371401bacc3f41678d5bfb82eb89e857097159b95715d7521f4b",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by-nc-sa/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2026-04-04T17:30:35Z",
    "title_canon_sha256": "d214cd220beeda91852a4ea93e3a7195d797345c2a044b3aa19e0ff6c26949ac"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.12522",
    "kind": "arxiv",
    "version": 1
  }
}