Pith Number

pith:B4GAKXG7

pith:2026:B4GAKXG7DDJLHSLGMF6VF7ZWLT

not attested not anchored not stored refs resolved

Encoding Robust Topological Signatures for Hyperdimensional Computing

Arpan Kusari

Topology-guided hyperdimensional computing resists image corruptions by encoding holes and rotation-invariant shape signatures.

arxiv:2605.16785 v1 · 2026-05-16 · 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{B4GAKXG7DDJLHSLGMF6VF7ZWLT}

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

Topology-guided HD computing substantially improves robustness compared with a naive HD baseline, maintaining high accuracy across multiple corruption families and benefiting from lightweight online training. Compared with a compact CNN trained on clean data, our method achieves competitive clean accuracy while offering markedly stronger robustness to several pixel-level corruptions.

C2weakest assumption

That holes and other topological primitives can be reliably extracted from binarized shapes even under the tested corruptions (rotation, Gaussian noise, salt-and-pepper, cutout, zoom) and that the chosen RTS-invariant descriptors (spatial-pyramid Zernike and intrinsic Fourier radial signatures) preserve the information needed for discrimination.

C3one line summary

Topology-guided HD computing encodes discrete holes and RTS-invariant descriptors (Zernike for outer shape, Fourier for holes) into hypervectors with learned reliability weights, yielding substantially higher robustness on corrupted MNIST/EMNIST than naive HD baselines while matching compact CNNs on

References

18 extracted · 18 resolved · 0 Pith anchors

[1] Frontiers in big data , volume= 2024

[2] 2025 11th International Conference on Computing and Artificial Intelligence (ICCAI) , pages= 2025

[3] Journal of Artificial Intelligence Research , volume=

[4] Cognitive computation , volume= 2009

[5] ACM Computing Surveys , volume= 2023

Formal links

1 machine-checked theorem link

Receipt and verification

First computed	2026-05-20T00:03:21.872417Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

0f0c055cdf18d2b3c966617d52ff365cfd0f64fe8309d83cb520c026aa97c83f

Aliases

arxiv: 2605.16785 · arxiv_version: 2605.16785v1 · doi: 10.48550/arxiv.2605.16785 · pith_short_12: B4GAKXG7DDJL · pith_short_16: B4GAKXG7DDJLHSLG · pith_short_8: B4GAKXG7

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/B4GAKXG7DDJLHSLGMF6VF7ZWLT \
  | 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: 0f0c055cdf18d2b3c966617d52ff365cfd0f64fe8309d83cb520c026aa97c83f

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "d4cd5c7cf2470a773e15050012182d663b3a16853b08463fa2822178d2756894",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CV",
    "submitted_at": "2026-05-16T03:36:40Z",
    "title_canon_sha256": "d33c6fa9025dccc531202d4bce85d9411fb1cb65975743e14f3c191da3cdd314"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.16785",
    "kind": "arxiv",
    "version": 1
  }
}