Lune

ACL Findings2026

From Signal Degradation to Computation Collapse: Uncovering the Two Failure Modes of LLM Quantization

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

Abstract

Post-Training Quantization (PTQ) is critical for the efficient deployment of Large Language Models (LLMs). While 4-bit quantization is widely regarded as an optimal trade-off, reducing the precision to 2-bit usually triggers a catastrophic "performance cliff." It remains unclear whether the underlying mechanisms differ fundamentally. Consequently, we conduct a systematic mechanistic analysis, revealing two qualitatively distinct failure modes: Signal Degradation, where the computational patterns remain intact but information precision is impaired by cumulative error; and Computation Collapse, where key components fail to function, preventing correct information processing and destroying the signal in the early layers. Guided by this diagnosis, we conduct mechanism-aware interventions, demonstrating that targeted, training-free repair can mitigate Signal Degradation, but remains ineffective for Computation Collapse. Our findings provide a systematic diagnostic framework for PTQ failures and suggest that addressing Computation Collapse requires structural reconstruction rather than mere compensation.

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

  • SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression cites SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models
  • SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models cites Measuring Massive Multitask Language Understanding
Builds onCited byLeft to right by year; the more cited, the higher.

Builds on12

  1. Measuring Massive Multitask Language UnderstandingDan Hendrycks, Collin Burns, Steven Basart, Andy Zou et al.ICLR 2021 · 7,905 citations
  2. Locating and Editing Factual Associations in GPTKevin Meng, David Bau, Alex Andonian, Yonatan BelinkovNeurIPS 2022 · 3,415 citations
  3. SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language ModelsGuangxuan Xiao, Ji Lin, Mickaël Seznec, Hao Wu et al.ICML 2023 · 1,493 citations
  4. HAWQ-V2: Hessian Aware trace-Weighted Quantization of Neural NetworksZhen Dong, Zhewei Yao, Daiyaan Arfeen, Amir Gholami et al.NeurIPS 2020 · 434 citations
  5. SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight CompressionTim Dettmers, Ruslan Svirschevski, Vage Egiazarian, Denis Kuznedelev et al.ICLR 2024 · 392 citations

Related papers

  1. 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
  2. Layer-Wise High-Impact Parameter Ratio Optimization in Post-Training Quantization for Large Language ModelsCuong Pham, Dung Anh Hoang, Cuong C. Nguyen, Trung Le et al.ACL 2026
  3. BPDQ: Bit-Plane Decomposition Quantization on a Variable Grid for Large Language ModelsJunyu Chen, Jungang Li, Jing Xiong, Wenjie Wang et al.ICML 2026
  4. CBQ: Cross-Block Quantization for Large Language ModelsXin Ding, Xiaoyu Liu, Zhijun Tu, Yun Zhang et al.ICLR 2025 · 1 citation
  5. Scaling Laws for PrecisionTanishq Kumar, Zachary Ankner, Benjamin Frederick Spector, Blake Bordelon et al.ICLR 2025
Dusk over the sea between two cliffs drawn in fine vertical lines