cs.CLApr 21, 2026

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

Authors: Chenxi ZhouPengfei CaoJiang LiBohan YuJinyu YeJun ZhaoKang Liu

Organizations: School of Advanced Interdisciplinary Sciences, University of Chinese Academy of Sciences · The Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences · College of Computer Science, Inner Mongolia University · School of Artificial Intelligence, University of Chinese Academy of Sciences

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.

Explore similar work

CardsList
  1. Why Does Post-Training Quantization Work?

    Sep 10, 2026Yuxiang Chen, Michael Beyer, Jun Zhu +1Weight-Only QuantizationQuantization