Lune

ACL Findings2026

Task-Stratified Knowledge Scaling Laws for Post-Training Quantized Large Language Models

Chenxi Zhou, Pengfei Cao, Jiang Li, Bohan Yu, Jinyu Ye, Jun Zhao, Kang Liu

Abstract

Post-Training Quantization (PTQ) is a critical strategy for efficient Large Language Models (LLMs) deployment. However, existing scaling laws primarily focus on general performance, overlooking crucial fine-grained factors and how quantization differentially impacts diverse knowledge capabilities. To address this, we establish Task-Stratified Knowledge Scaling Laws. By stratifying capabilities into memorization, application, and reasoning, we develop a framework that unifies model size, bitwidth, and fine-grained factors: group size and calibration set size. Validated on 293 diverse PTQ configurations, our framework demonstrates strong fit and cross-architecture consistency. It reveals distinct sensitivities across knowledge capabilities: reasoning is precisioncritical, application is scale-responsive, and memorization is calibration-sensitive. We highlight that in low-bit scenarios, optimizing these fine-grained factors is essential for preventing performance collapse. These findings provide an empirically-backed foundation for designing knowledge-aware quantization strategies. Knowledge Reasoning Knowledge Reasoning Factual Recall Flexible Application Multi-step Reasoning

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

  • Evaluating Quantized Large Language Models cites Measuring Massive Multitask Language Understanding
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on12

  1. Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
  2. Measuring Massive Multitask Language UnderstandingDan Hendrycks, Collin Burns, Steven Basart, Andy Zou et al.ICLR 2021 · 7,905 citations
  3. WinoGrande: An Adversarial Winograd Schema Challenge at ScaleKeisuke Sakaguchi, Ronan Le Bras, Chandra Bhagavatula, Yejin ChoiAAAI 2020 · 3,037 citations
  4. QuIP: 2-Bit Quantization of Large Language Models With GuaranteesJerry Chee, Yaohui Cai, Volodymyr Kuleshov, Christopher De SaNeurIPS 2023 · 503 citations
  5. Evaluating Quantized Large Language ModelsShiyao Li, Xuefei Ning, Luning Wang, Tengxuan Liu et al.ICML 2024 · 88 citations

Related papers

  1. Towards Quantization-Aware Training for Ultra-Low-Bit Reasoning LLMsYasuyuki Okoshi, Hikari Otsuka, Daichi Fujiki, Masato MotomuraICLR 2026
  2. FAQ: Mitigating Quantization Error via Regenerating Calibration Data with Family-Aware QuantizationHaiyang Xiao, Weiqing Li, Jinyue Guo, Guochao Jiang et al.ACL Findings 2026
  3. MLWQ: Efficient Small Language Model Deployment via Multi-Level Weight QuantizationChun Hu, Junhui He, Shangyu Wu, Yuxin He et al.EMNLP 2025 · 1 citation
  4. SliM-LLM: Salience-Driven Mixed-Precision Quantization for Large Language ModelsWei Huang, Haotong Qin, Yangdong Liu, Yawei Li et al.ICML 2025
  5. SEPTQ: A Simple and Effective Post-Training Quantization Paradigm for Large Language ModelsHan Liu, Haotian Gao, Xiaotong Zhang, Changya Li et al.KDD 2025 · 1 citation
Dusk over the sea between two cliffs drawn in fine vertical lines