Pith Number

pith:WIPZ73I5

pith:2026:WIPZ73I52ME4SRCRR4QEP5SBSZ

not attested not anchored not stored refs resolved

FLASH: Efficient Visuomotor Policy via Sparse Sampling

Gen Li, Jianfei Yang, Jiaqi Bai, Jindou Jia, Kuangji Zuo, Tuo An, Xiangyu Chen, Yuxuan Hu

A visuomotor policy using Legendre polynomials and history-anchored flow matching generates long robot action sequences in a single fast step.

arxiv:2605.15492 v1 · 2026-05-15 · cs.RO · cs.CV

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{WIPZ73I52ME4SRCRR4QEP5SBSZ}

Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more

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

FLASH achieves state-of-the-art success rates (>=92% across all tasks), a per-episode inference time of 31.40 ms (up to 175x faster than diffusion policies and 18x faster than prior flow matching policies), up to 4x faster training convergence than ACT, and 5x to 7x reduction in controller tracking error compared to discrete-action baselines.

C2weakest assumption

That fitting expert demonstrations under sparse temporal sampling combined with initialization from history polynomial coefficients enables accurate single-step flow matching that preserves performance over extended action horizons without post-hoc tuning or task-specific adjustments.

C3one line summary

FLASH Policy uses sparse Legendre polynomial trajectory fitting and history-anchored flow matching to enable single-step inference for visuomotor control, reporting 31.4 ms per-episode latency and >=92% success on five simulated plus two real manipulation tasks.

References

33 extracted · 33 resolved · 6 Pith anchors

[1] Advances in Neural Information Processing Systems , volume=

[2] Proceedings of International Conference on Learning Representations , year=

[3] Black, Kevin and Brown, Noah and Driess, Danny and Esmail, Adnan and Equi, Michael and Finn, Chelsea and Fusai, Niccolo and Groom, Lachy and Hausman, Karol and Ichter, Brian and others , journal=

[4] The International Journal of Robotics Research , volume= 2025

[5] Denoising Diffusion Implicit Models 2010 · arXiv:2010.02502

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-20T00:01:01.473024Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

b21f9fed1dd309c944518f2047f6419644063e49994aa582aea0eb6152670575

Aliases

arxiv: 2605.15492 · arxiv_version: 2605.15492v1 · doi: 10.48550/arxiv.2605.15492 · pith_short_12: WIPZ73I52ME4 · pith_short_16: WIPZ73I52ME4SRCR · pith_short_8: WIPZ73I5

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/WIPZ73I52ME4SRCRR4QEP5SBSZ \
  | 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: b21f9fed1dd309c944518f2047f6419644063e49994aa582aea0eb6152670575

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "9b2d27316ecf7e6e220612f986bece989b2a13fd2ba0bb80c8c43bf5143aaf03",
    "cross_cats_sorted": [
      "cs.CV"
    ],
    "license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
    "primary_cat": "cs.RO",
    "submitted_at": "2026-05-15T00:15:24Z",
    "title_canon_sha256": "7c5d5f36ffc93540fbb55d525aa4545c95ffa9220da490c37e37c1dfe7b2d483"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.15492",
    "kind": "arxiv",
    "version": 1
  }
}