From Sweep to Seam: Interleaved Cross-Block Post-Training Quantization

📅 2026-08-10
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
This work addresses the performance degradation in deep neural networks caused by error accumulation in conventional block-wise post-training quantization (PTQ), where a single forward pass prevents correction of early quantization errors. To mitigate this issue, the authors propose Interleaved Cross-Block Quantization (ICBQ), which refines local reconstruction by doubly optimizing adjacent “seam” block pairs at block boundaries. Built upon standard block-wise PTQ, ICBQ incorporates cross-block reconstruction within a sliding window, leveraging local shrinkage and smoothness assumptions, and is compatible with GPTQ scheduling. Experimental results demonstrate that ICBQ significantly reduces perplexity under ternary, 2-bit, and 3-bit quantization, effectively preventing performance collapse at extreme compression ratios while seamlessly integrating into existing PTQ pipelines.
📝 Abstract
Compressing large language models to two bits or fewer is increasingly feasible through block-wise post-training quantization; cross-block variants reconstruct neighboring Transformer blocks within a moving window. In the fixed two-block setting studied here, the matched sequential baseline moves this window through the network once, so errors introduced early in the sweep are not revisited. We propose Interleaved Cross-Block Quantization (ICBQ), a scheduling modification that revisits the boundary pair between consecutive chunks. Each seam pair is refined twice: first at the end of one chunk and again at the start of the next. The method retains the local two-block objective and reuses the calibration inputs of existing block-wise PTQ pipelines. Under stated local contraction and smoothness assumptions, we derive a depth-wise upper-bound comparison in which seam revisits multiply the propagated term while the residual remains bounded independently of depth. In the reported experiments, ICBQ reduces ternary-quantization perplexity relative to the matched Sequential CBQ baseline, yields finite perplexity in configurations where the baseline has severe degradation, and can also be used with 3-bit and 2-bit GPTQ.
Problem

Research questions and friction points this paper is trying to address.

post-training quantization
large language models
cross-block quantization
error propagation
model compression
Innovation

Methods, ideas, or system contributions that make the work stand out.

Interleaved Cross-Block Quantization
Post-Training Quantization
Large Language Model Compression
Error Propagation Control
Seam Refinement
🔎 Similar Papers
💼 Related Jobs
No related jobs found.
A
Achille Jacquemond
Fujitsu Limited
Y
Yuma Ichikawa
Fujitsu Limited, RIKEN Center for AIP
A
Akira Sakai
Fujitsu Limited, Tokai University