Pith Number

pith:L2NAT4AJ

pith:2026:L2NAT4AJVQOD4QLPBPKNMN7JEW

not attested not anchored not stored refs resolved

Realtime-VLA FLASH: Speculative Inference Framework for Diffusion-based VLAs

Huawei Li, Jiahui Niu, Kefan Gu, Shengwen Liang, Tiancai Wang, Xing Hu, Ying Wang, Yucheng Zhao

A lightweight draft model with parallel verification lets diffusion VLAs replan actions at 19.1 ms average latency instead of 58 ms.

arxiv:2605.13778 v1 · 2026-05-13 · cs.RO · 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{L2NAT4AJVQOD4QLPBPKNMN7JEW}

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

Experiments show that on LIBERO, FLASH largely preserves task performance by replacing many 58.0 ms full-inference rounds with speculative rounds as fast as 7.8 ms, lowering task-level average inference latency to 19.1 ms (3.04x speedup).

C2weakest assumption

The lightweight draft model plus parallel Action Expert verification and phase-aware fallback will trigger full inference infrequently enough to deliver net speedup while keeping task success rates intact across varied environments.

C3one line summary

A new speculative inference system speeds up diffusion VLAs to 19.1 ms average latency (3.04x faster) on LIBERO by replacing most full 58 ms inferences with 7.8 ms draft rounds while preserving task performance.

References

35 extracted · 35 resolved · 15 Pith anchors

[1] $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control 2024 · arXiv:2410.24164

[2] Real-Time Execution of Action Chunking Flow Policies 2025 · arXiv:2506.07339

[3] Ren, Michael Equi, and Sergey Levine 2025

[4] Mean-flow based one-step vision-language-action 2026

[5] Diffusion policy: Visuomotor policy learning via action diffusion 2023

Receipt and verification

First computed	2026-05-18T02:44:15.858263Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

5e9a09f009ac1c3e416f0bd4d637e925bad7597b9474f0bcb0ce5c92d3b473f1

Aliases

arxiv: 2605.13778 · arxiv_version: 2605.13778v1 · doi: 10.48550/arxiv.2605.13778 · pith_short_12: L2NAT4AJVQOD · pith_short_16: L2NAT4AJVQOD4QLP · pith_short_8: L2NAT4AJ

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/L2NAT4AJVQOD4QLPBPKNMN7JEW \
  | 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: 5e9a09f009ac1c3e416f0bd4d637e925bad7597b9474f0bcb0ce5c92d3b473f1

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "33e9e93ceefba81c0efa24a018b8e9fa2a019351c5f55fa5e3497bdbc7cdf829",
    "cross_cats_sorted": [
      "cs.CV"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.RO",
    "submitted_at": "2026-05-13T16:57:51Z",
    "title_canon_sha256": "d297bc85bf7f8cfc6c64f5c6e7cbc13e5168f05dcf5fbabb555b53460a8cbbd4"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.13778",
    "kind": "arxiv",
    "version": 1
  }
}