pith:P6FOLOHK
ReSpinQuant: Efficient Layer-Wise LLM Quantization via Subspace Residual Rotation Approximation
ReSpinQuant approximates per-layer rotation matrices with residual subspaces so that layer-wise LLM quantization accuracy can be obtained at the speed of global rotation methods.
arxiv:2604.11080 v2 · 2026-04-13 · cs.CV · cs.AI
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{P6FOLOHKQF3CJ63EFUGURZ5M5J}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
ReSpinQuant resolves such overhead by leveraging offline activation rotation fusion and matching basis using efficient residual subspace rotation. This design reconciles the high expressivity of layer-wise adaptation with only negligible inference overhead.
That the residual subspace rotation approximation can capture enough of the expressivity of full per-layer transformations to match their accuracy while still permitting complete offline fusion into the model weights.
ReSpinQuant achieves state-of-the-art accuracy in W4A4 and W3A3 LLM quantization by using efficient residual subspace rotation approximations that match layer-wise performance while retaining the inference speed of global rotation methods.
Receipt and verification
| First computed | 2026-05-29T01:05:09.286260Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
7f8ae5b8ea817624fb642d0d48e7acea6ba25439de903ad1f0ad9a957fcb0c23
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/P6FOLOHKQF3CJ63EFUGURZ5M5J \
| 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: 7f8ae5b8ea817624fb642d0d48e7acea6ba25439de903ad1f0ad9a957fcb0c23
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "6d7e911cf95b267516cc6d47ae2623387979e2ad3036e33746ae12846aff88b7",
"cross_cats_sorted": [
"cs.AI"
],
"license": "http://creativecommons.org/licenses/by/4.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-04-13T07:00:26Z",
"title_canon_sha256": "2715ada39e29f0897f43ea0a97b9a20d988d503534cf45afe4a4a47a39ff3a29"
},
"schema_version": "1.0",
"source": {
"id": "2604.11080",
"kind": "arxiv",
"version": 2
}
}