Pith Number

pith:XDHKGM5I

pith:2026:XDHKGM5IIL27N6LIIMDHYGDGKD

not attested not anchored not stored refs resolved

{\mu}-ORCA: Optimizing Acceleration for Microsecond-Scale Deep Neural Network Inference on ACAP

Jinming Zhuang, Peipei Zhou, Shixin Ji, Wei Zhang, Xingzhen Chen, Zhuoping Yang

μ-ORCA achieves 0.93 μs DNN inference latency on ACAP by direct AIE array communication

arxiv:2605.17683 v1 · 2026-05-17 · cs.AR

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{XDHKGM5IIL27N6LIIMDHYGDGKD}

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

μ-ORCA achieves 0.93 μs latency for a 6-layer real-world DeepSets model, satisfying the 1-μs latency budget, with average latency reduction of >1.70× and >1.83× compared with different state-of-the-art ACAP frameworks.

C2weakest assumption

The assumption that direct inter-layer communication on the AIE array can be realized with negligible additional synchronization or routing overhead for the small problem sizes typical of jet-tagging models (section on framework design and evaluation).

C3one line summary

μ-ORCA achieves 0.93 μs end-to-end latency for a 6-layer DeepSets model on AMD ACAP VEK280 by direct AIE-array inter-layer communication and overhead-aware design space exploration, delivering over 1.7× latency reduction versus prior ACAP frameworks.

References

38 extracted · 38 resolved · 1 Pith anchors

[1] Real time analysis with the CMS Level-1 Trigger, 2025 2025

[2] The next-generation triggers for CERN detectors, 2025 2025

[3] Taking a closer look at LHC, 2025 2025

[4] Abdelkhalik, H. et al. Demystifying the Nvidia Ampere Architecture through Microbenchmarking and Instruction-level Analysis. InHPEC, pages 1–8, 2022 2022

[5] Zhuang, J. et al. CHARM: Composing Heterogeneous AcceleRators for Matrix Multiply on Versal ACAP Architecture. InFPGA, New York, NY, USA, 2023 2023

Formal links

2 machine-checked theorem links

Receipt and verification

First computed	2026-05-20T00:04:52.577087Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

b8cea333a842f5f6f96843067c186650dd667fcc39be57a3d0be7840a1ece6e3

Aliases

arxiv: 2605.17683 · arxiv_version: 2605.17683v1 · doi: 10.48550/arxiv.2605.17683 · pith_short_12: XDHKGM5IIL27 · pith_short_16: XDHKGM5IIL27N6LI · pith_short_8: XDHKGM5I

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/XDHKGM5IIL27N6LIIMDHYGDGKD \
  | 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: b8cea333a842f5f6f96843067c186650dd667fcc39be57a3d0be7840a1ece6e3

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "17763b3a3a341386061c71ce32c4ecc596bdaae28f33ae62b33c7f6e909aafca",
    "cross_cats_sorted": [],
    "license": "http://creativecommons.org/licenses/by-nc-nd/4.0/",
    "primary_cat": "cs.AR",
    "submitted_at": "2026-05-17T22:54:01Z",
    "title_canon_sha256": "0064039c165b2171fcad1188bf5eec90ced4551ff0a7180646813ae09f3db816"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.17683",
    "kind": "arxiv",
    "version": 1
  }
}