Pith Number

pith:BOVYFUXR

pith:2026:BOVYFUXRPEGMZCOI22DN4CQVBB

not attested not anchored not stored refs resolved

CoReDiT: Spatial Coherence-Guided Token Pruning and Reconstruction for Efficient Diffusion Transformers

Fatih Porikli, Hong Cai, Hsin-Pai Cheng, Shizhong Han, Zhuojin Li

CoReDiT prunes redundant tokens in diffusion transformers via spatial coherence scores and reconstructs their outputs from neighbors to cut computation while preserving quality.

arxiv:2605.14191 v1 · 2026-05-13 · cs.CV

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{BOVYFUXRPEGMZCOI22DN4CQVBB}

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

Across state-of-the-art diffusion backbones including PixArt-α and MagicDrive-V2, CoReDiT achieves up to 55% self-attention FLOPs reduction and inference speedups of 1.33x on cloud GPUs and 1.72x on mobile NPUs, while maintaining high visual quality.

C2weakest assumption

That the linear-time spatial coherence score reliably identifies redundant tokens whose removal and subsequent neighbor-based reconstruction will not introduce perceptible visual artifacts or degrade generation quality across diverse prompts and resolutions.

C3one line summary

CoReDiT reduces self-attention FLOPs in DiTs by up to 55% via linear-time spatial coherence pruning and neighbor-based reconstruction, delivering 1.33x-1.72x speedups with maintained quality.

References

39 extracted · 39 resolved · 3 Pith anchors

[1] Token merging for fast sta- ble diffusion

[2] Token Merging: Your ViT But Faster 2022 · arXiv:2210.09461

[3] nuscenes: A multi- modal dataset for autonomous driving 2020

[4] Exploring diffusion transformer designs via grafting 2025

[5] Flexdit: Dynamic token density control for diffusion transformer, 2024 2024

Receipt and verification

First computed	2026-05-17T23:39:11.140095Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

0bab82d2f1790ccc89c8d686de0a15087affee2a7a4c89d224d2f297efb2f2c6

Aliases

arxiv: 2605.14191 · arxiv_version: 2605.14191v1 · doi: 10.48550/arxiv.2605.14191 · pith_short_12: BOVYFUXRPEGM · pith_short_16: BOVYFUXRPEGMZCOI · pith_short_8: BOVYFUXR

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/BOVYFUXRPEGMZCOI22DN4CQVBB \
  | 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: 0bab82d2f1790ccc89c8d686de0a15087affee2a7a4c89d224d2f297efb2f2c6

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "d636a1122738d4a1e98a66b91ae9bd1f4347dc86445d0a5f89793b45c77691d3",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-13T23:13:29Z",
    "title_canon_sha256": "5f3f341b8d95770267ac541fa599d623099a1bfc01bbbc1bed784f96d6c05e82"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.14191",
    "kind": "arxiv",
    "version": 1
  }
}