pith. sign in
Pith Number

pith:46ATTTLZ

pith:2026:46ATTTLZ3TGGDKJHWKXFEUBYZ6
not attested not anchored not stored refs resolved

Fitting Is Not Enough: Smoothness in Extremely Quantized LLMs

Pengzhan Li, Wanxiang Che, Xu Han, Yuxuan Li, Yuzhuang Xu

Extremely quantized LLMs lose generation quality from smoothness degradation in token predictions, beyond numerical accuracy loss alone.

arxiv:2605.08894 v2 · 2026-05-09 · cs.CL · cs.AI

Add to your LaTeX paper
\usepackage{pith}
\pithnumber{46ATTTLZ3TGGDKJHWKXFEUBYZ6}

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

Extremely quantized LLMs suffer from systematic smoothness degradation beyond numerical precision loss; preserving smoothness via a simple principle in post-training quantization and quantization-aware training brings additional gains beyond numerical accuracy.

C2weakest assumption

The smoothness proxy used in the paper accurately captures the degradation that matters for generation quality, and the observed reduction in effective token candidates is a direct causal consequence of that smoothness loss rather than an artifact of the proxy or the evaluation setup.

C3one line summary

Extremely quantized LLMs exhibit systematic smoothness degradation that reduces effective token candidates and degrades generation; a smoothness-preserving principle in PTQ and QAT delivers gains beyond numerical accuracy.

References

46 extracted · 46 resolved · 8 Pith anchors

[1] A. Amini, S. Gabriel, S. Lin, R. Koncel-Kedziorski, Y . Choi, and H. Hajishirzi. MathQA: Towards interpretable math word problem solving with operation-based formalisms. InProceed- ings of the 2019 Co 2019 · doi:10.18653/v1/n19-1245
[2] D. Bahri, H. Mobahi, and Y . Tay. Sharpness-aware minimization improves language model generalization. InProceedings of the 60th Annual Meeting of the Association for Computational Linguistics (ACL), 2022 · doi:10.18653/v1/2022
[3] P. L. Bartlett, D. J. Foster, and M. J. Telgarsky. Spectrally-normalized margin bounds for neural networks.Advances in neural information processing systems (NeurIPS), 30:6240–6249, 2017. URL https:// 2017
[4] PIQA: Reasoning about Physical Commonsense in Natural Language 2020 · doi:10.48550/arxiv.1911.11641
[5] A. Chan, Y . Tay, and Y .-S. Ong. What it thinks is important is important: Robustness transfers through input gradients. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recogn 2020

Formal links

2 machine-checked theorem links

Receipt and verification
First computed 2026-05-20T00:00:41.694431Z
Builder pith-number-builder-2026-05-17-v1
Signature Pith Ed25519 (pith-v1-2026-05) · public key
Schema pith-number/v1.0

Canonical hash

e78139cd79dccc61a927b2ae525038cf8292447e1628b3a8c823407e2577ce60

Aliases

arxiv: 2605.08894 · arxiv_version: 2605.08894v2 · doi: 10.48550/arxiv.2605.08894 · pith_short_12: 46ATTTLZ3TGG · pith_short_16: 46ATTTLZ3TGGDKJH · pith_short_8: 46ATTTLZ
Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/46ATTTLZ3TGGDKJHWKXFEUBYZ6 \
  | 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: e78139cd79dccc61a927b2ae525038cf8292447e1628b3a8c823407e2577ce60
Canonical record JSON
{
  "metadata": {
    "abstract_canon_sha256": "5d517aa534d4cec95f2c9559f81d0e05402905774f9862b7e56e0368fc6a989f",
    "cross_cats_sorted": [
      "cs.AI"
    ],
    "license": "http://creativecommons.org/licenses/by/4.0/",
    "primary_cat": "cs.CL",
    "submitted_at": "2026-05-09T11:19:51Z",
    "title_canon_sha256": "9ff425cadc9833ff754fc12594764fc53634ceadec8ea2ff6100303586bf7cfe"
  },
  "schema_version": "1.0",
  "source": {
    "id": "2605.08894",
    "kind": "arxiv",
    "version": 2
  }
}