Pith Number

pith:QDWH4FYS

pith:2026:QDWH4FYSKHFSJLVREWN4PU27N2

not attested not anchored not stored refs resolved

HASTE: Training-Free Video Diffusion Acceleration via Head-Wise Adaptive Sparse Attention

Fei Chao, Jing Xu, Rongrong Ji, Xiawu Zheng, Xuzhe Zheng, Yuexiao Ma

Head-wise adaptive sparse attention accelerates pretrained video diffusion models up to 1.93 times without retraining by reusing temporal masks and calibrating sparsity per head.

arxiv:2605.14513 v1 · 2026-05-14 · cs.CV · 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{QDWH4FYSKHFSJLVREWN4PU27N2}

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 Wan2.1-1.3B and Wan2.1-14B, our method consistently improves XAttention and SVG2, achieving up to 1.93 times speedup at 720P while maintaining competitive video quality and similarity metrics.

C2weakest assumption

That measured model-output error under a global sparsity budget reliably predicts perceptual video quality across heads and that temporal query-key drift is stable enough for safe mask reuse without visible artifacts.

C3one line summary

HASTE delivers up to 1.93x speedup on Wan2.1 video DiTs via head-wise adaptive sparse attention using temporal mask reuse and error-guided per-head calibration while preserving video quality.

References

55 extracted · 55 resolved · 11 Pith anchors

[1] Dicache: Let diffusion model determine its own cache 2025

[2] Rainfusion: Adaptive video generation acceleration via multi-dimensional visual redundancy 2025

[3] RainFusion2.0: Temporal-Spatial Awareness and Hardware-Efficient Block-wise Sparse Attention 2025 · arXiv:2512.24086

[4] Sparse-vDiT: Unleashing the Power of Sparse Attention to Accelerate Video Diffusion Transformers, June 2025 2025

[5] Hicache: Training-free acceleration of diffusion models via hermite polynomial- based feature caching.arXiv preprint arXiv:2508.16984 2025

Receipt and verification

First computed	2026-05-17T23:39:06.168512Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

80ec7e171251cb24aeb1259bc7d35f6e837a4669680ba7991c4b865869cf8992

Aliases

arxiv: 2605.14513 · arxiv_version: 2605.14513v1 · doi: 10.48550/arxiv.2605.14513 · pith_short_12: QDWH4FYSKHFS · pith_short_16: QDWH4FYSKHFSJLVR · pith_short_8: QDWH4FYS

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/QDWH4FYSKHFSJLVREWN4PU27N2 \
  | 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: 80ec7e171251cb24aeb1259bc7d35f6e837a4669680ba7991c4b865869cf8992

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "f04225a61e43de816dd86645f93e02d3bb37c9e5758932130e81af872c72a596",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-14T07:57:55Z",
    "title_canon_sha256": "b9227144d45ce8fd31aefe215632180d3954b4c277575990c45d9044dae0a580"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.14513",
    "kind": "arxiv",
    "version": 1
  }
}