Pith Number

pith:CKYPY7CO

pith:2026:CKYPY7COKENL7RYMHMYDYQI2JZ

not attested not anchored not stored refs resolved

Precise Verification of Transformers through ReLU-Catalyzed Abstraction Refinement

Hengjie Liu, Jianjun Zhao, Zhenya Zhang

ReLU encoding of dot-product ranges enables tighter convex bounds for precise transformer verification.

arxiv:2605.14294 v1 · 2026-05-14 · cs.AI · cs.LG

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

Add to your LaTeX paper

\usepackage{pith}
\pithnumber{CKYPY7COKENL7RYMHMYDYQI2JZ}

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

we propose a transformer verification approach that can achieve improved precision... by representing a precise but non-linear bound for dot products such that we can further exploit the rich body of literature for convex relaxation of ReLU to derive precise bounds.

C2weakest assumption

That the ReLU encoding of dot-product ranges remains tractable for convex relaxation and that the resulting bounds are tight enough to meaningfully reduce false alarms on the evaluated model sizes and properties.

C3one line summary

A ReLU-catalyzed abstraction method yields tighter bounds for transformer verification by converting dot-product constraints into ReLU forms that leverage standard convex relaxations.

References

39 extracted · 39 resolved · 2 Pith anchors

[1] Alzantot, M., Sharma, Y., Elgohary, A., Ho, B.J., Srivastava, M., Chang, K.W.: Generating natural language adversarial examples. In: Riloff, E., Chiang, D., Hocken- maier, J., Tsujii, J. (eds.) Procee 2018 · doi:10.18653/v1/d18-1316

[2] In: Sedoc, J., Rogers, A., Rumshisky, A., Tafreshi, S 2021 · doi:10.18653/v1/2021.insights-1.18

[3] Lee, and Brandon Reagen 2021 · doi:10.1145/3453483.3454056

[4] Boudardara, F., Boussif, A., Meyer, P.J., Ghazel, M.: A review of abstraction methods toward verifying neural networks. ACM Trans. Embed. Comput. Syst. 23(4) (Jun 2024). https://doi.org/10.1145/361750 2024 · doi:10.1145/3617508

[5] URLhttps://doi.org/10.48550/arXiv 2024 · doi:10.48550/arxiv

Receipt and verification

First computed	2026-05-17T23:39:10.168048Z
Builder	pith-number-builder-2026-05-17-v1
Signature	Pith Ed25519 (`pith-v1-2026-05`) · public key
Schema	pith-number/v1.0

Canonical hash

12b0fc7c4e511abfc70c3b303c411a4e4ab8922d463e89f04b20551c2257ba9d

Aliases

arxiv: 2605.14294 · arxiv_version: 2605.14294v1 · doi: 10.48550/arxiv.2605.14294 · pith_short_12: CKYPY7COKENL · pith_short_16: CKYPY7COKENL7RYM · pith_short_8: CKYPY7CO

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/CKYPY7COKENL7RYMHMYDYQI2JZ \
  | 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: 12b0fc7c4e511abfc70c3b303c411a4e4ab8922d463e89f04b20551c2257ba9d

Canonical record JSON

{
  "metadata": {
    "abstract_canon_sha256": "b492531e79708fa65f0e9c1cb0ff89d73220ad2b7dad08d8858f54327095a83e",
    "cross_cats_sorted": [
      "cs.LG"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.AI",
    "submitted_at": "2026-05-14T02:55:53Z",
    "title_canon_sha256": "e95d0dd3b25c1ed57ab754e0f7e89c430dd56366b72c4bcbe89f84d95ffec567"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.14294",
    "kind": "arxiv",
    "version": 1
  }
}