When Flat Minima Fail: Characterizing INT4 Quantization Collapse After FP32 Convergence
Post-training quantization (PTQ) assumes that a well-converged model is a quantization-ready model. We show this assumption fails in a structured, measurable, and previously uncharacterized way. Using a calibration-free per-group INT4 probe applied to all 154 publicly available Pythia-160m training checkpoints, we identify a three-phase divergence structure: a rapid-learning phase where both FP32 perplexity and quantization robustness improve together, a meta-stable plateau lasting roughly 70,000 steps where FP32 perplexity stagnates but INT4 gap remains bounded, and an explosive divergence phase where the INT4 gap compounds from 11% to 517% while FP32 perplexity barely moves. Critically, this divergence begins not when the learning rate starts decaying, but precisely when FP32 perplexity converges a finer-grained onset predictor that implies post-convergence weight updates, rather than decay magnitude alone, are the proximate cause. We further show that INT8 quantization is entirely immune throughout all three phases, constraining the mechanism to the coarseness of the 16-level INT4 grid specifically, and rule out weight outlier accumulation as the mechanism via direct kurtosis measurement. Finally, we conduct a controlled fork experiment from the pre-divergence checkpoint comparing three learning rate schedules (cosine continuation, SGDR warm restarts, and our proposed Oscillatory Lock-In) across nine independent runs. SGDR uniformly accelerates divergence (0/9 pairwise wins against cosine), while OLI's settled cool phases reduce the INT4 gap by 2.2 percentage points on average (t = -5.46, p < 0.0001), demonstrating that schedule amplitude calibration, not oscillation alone, determines whether perturbation helps or hurts. Our code, probe implementation, and all 154-checkpoint audit results are released publicly.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
QT-DoG: Quantization-aware Training for Domain Generalization
Domain Generalization (DG) aims to train models that perform well not only on the training (source) domains but also on novel, unseen target data distributions. A key challenge in DG is preventing overfitting to source d…
Domain GeneralizationModel CompressionQuantizationSQuAT: Sharpness- and Quantization-Aware Training for BERT
Quantization is an effective technique to reduce memory footprint, inference latency, and power consumption of deep learning models. However, existing quantization methods suffer from accuracy degradation compared to ful…
QuantizationGAQAT: gradient-adaptive quantization-aware training for domain generalization
Research on loss surface geometry, such as Sharpness-Aware Minimization (SAM), shows that flatter minima improve generalization. Recent studies further reveal that flatter minima can also reduce the domain generalization…
Domain GeneralizationQuantizationUnderstanding Flatness in Generative Models: Its Role and Benefits
Flat minima, known to enhance generalization and robustness in supervised learning, remain largely unexplored in generative models. In this work, we systematically investigate the role of loss surface flatness in generat…
Noise EstimationQuantizationFlattenQuant: Breaking Through the Inference Compute-bound for Large Language Models with Per-tensor Quantization
Large language models (LLMs) have demonstrated state-of-the-art performance across various tasks. However, the latency of inference and the large GPU memory consumption of LLMs restrict their deployment performance. Rece…
GPUQuantization