pith:MQB2XUF2
PACZero: PAC-Private Fine-Tuning of Language Models via Sign Quantization
PACZero achieves usable fine-tuning performance for large language models at zero mutual information privacy by using sign quantization to create unanimous gradient updates.
arxiv:2605.06505 v2 · 2026-05-07 · cs.LG · cs.AI · cs.CR
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{MQB2XUF2N7GH5GOMCZBVGVGCPA}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
On SST-2 OPT-1.3B full fine-tuning at I=0, PACZero-ZPL reaches 88.99±0.91, within 2.1pp of the non-private MeZO baseline (91.1 FT). No prior method produces usable utility in the high-privacy regime ε<1.
That sign-quantizing subset-aggregated zeroth-order gradients creates frequent unanimity steps at which every candidate subset agrees on the update direction, making the released sign cost zero conditional mutual information.
PACZero achieves zero mutual information privacy for LLM fine-tuning via sign-quantized zeroth-order gradients, delivering near-non-private accuracy on SST-2 and SQuAD at I=0.
Receipt and verification
| First computed | 2026-05-26T02:04:12.314302Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
6403abd0ba6fcc7e99cc16435354c2783006a832cb9f145f687c8e63a84fefa9
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/MQB2XUF2N7GH5GOMCZBVGVGCPA \
| 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: 6403abd0ba6fcc7e99cc16435354c2783006a832cb9f145f687c8e63a84fefa9
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "10ea4d6835e476e66b872d0bd2e2c77dc0b4de83eeb0609c1f40fcf168c2e341",
"cross_cats_sorted": [
"cs.AI",
"cs.CR"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"primary_cat": "cs.LG",
"submitted_at": "2026-05-07T16:20:20Z",
"title_canon_sha256": "4afda9e9e538af6078b31b61b06534fd7fd1fe29c57341832d6b815c2db4a9aa"
},
"schema_version": "1.0",
"source": {
"id": "2605.06505",
"kind": "arxiv",
"version": 2
}
}