Pith Number

pith:J7DZSRRH

pith:2026:J7DZSRRHZSUK4DDTYEUNXC7YTX

not attested not anchored not stored refs resolved

What Concepts Lie Within? Detecting and Suppressing Risky Content in Diffusion Transformers

An-An Liu, Chenyu Zhang, Lanjun Wang, Ruidong Chen, Wenhui Li, Yueyang Cheng

Attention heads in diffusion transformers show concept-specific sensitivity that lets risky content be detected and suppressed at inference time.

arxiv:2605.10180 v1 · 2026-05-11 · cs.CV · cs.CR

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{J7DZSRRHZSUK4DDTYEUNXC7YTX}

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

attention heads exhibit concept-specific sensitivity. This property enables both the detection and suppression of risky content.

C2weakest assumption

The assumption that concept-specific sensitivity in attention heads is stable enough across prompts and models to allow reliable detection and suppression of risky tokens without degrading overall image quality or being easily circumvented.

C3one line summary

A method using attention head vectors detects and suppresses risky content generation in Diffusion Transformers at inference time.

References

51 extracted · 51 resolved · 6 Pith anchors

[1] Praneeth Bedapudi. 2019. NudeNet: Neural Nets for Nudity Detection and Cen- soring. https://github.com/bedapudi6788/NudeNet. Python package, version 1.1.0 2019

[2] Black Forest Labs. 2025. FLUX.1 [dev] — Model Card. https://huggingface.co/ black-forest-labs/FLUX.1-dev 2025

[3] Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer 2025 · arXiv:2511.22699

[4] Qi Cai, Yehao Li, Yingwei Pan, Ting Yao, and Tao Mei. 2025. HiDream-I1: An Open-Source High-Efficient Image Generative Foundation Model. InProceedings of the 33rd ACM International Conference on Multi 2025

[5] PixArt-$\alpha$: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis 2023 · arXiv:2310.00426

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-20T17:03:46.339984Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

4fc7994627cca8ae0c73c128db8bf89dd91e96ae013bf9d66196e948314b5872

Aliases

arxiv: 2605.10180 · arxiv_version: 2605.10180v1 · doi: 10.48550/arxiv.2605.10180 · pith_short_12: J7DZSRRHZSUK · pith_short_16: J7DZSRRHZSUK4DDT · pith_short_8: J7DZSRRH

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/J7DZSRRHZSUK4DDTYEUNXC7YTX \
  | 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: 4fc7994627cca8ae0c73c128db8bf89dd91e96ae013bf9d66196e948314b5872

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "7962f16007da89688f376f618fc6b8e422059591c07a1bc3255eb1f9bffea8c9",
    "cross_cats_sorted": [
      "cs.CR"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-11T08:31:57Z",
    "title_canon_sha256": "5bd83ebcd5a390aaa8f1b15e30b8a1f02c964df625bd043cfdba626a1c5caca1"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.10180",
    "kind": "arxiv",
    "version": 1
  }
}