QRazor: Reliable and Effortless 4-bit LLM Quantization by Significant Data Razoring
Keywords: Large Language Model, Post-Training Quantization, Hardware, Arithmetic Unit, Compression
TL;DR: QRazor introduces a reliable and effortless 4-bit quantization method that achieves competitive accuracy in large language models with reduced computational complexity.
Abstract: Large-scale language models (LLMs) have demonstrated outstanding performance in language processing tasks, yet their deployment is often hindered by high memory demands and computational complexity. Although low-bit quantization techniques, such as 4-bit quantization, present a potential solution, they frequently lead to significant accuracy degradation or require substantial effort for such aggressive quantization approaches. To overcome these challenges, we introduce QRazor, a reliable and effortless quantization scheme designed to enable 4-bit quantization for weights, activations, and KV cache in transformer-based LLMs. The scheme involves two main stages: quantization and compression. During the quantization stage, weights, activations, and KV cache values are quantized with wider 8 or 16-bit integers as a basis to achieve nearly identical accuracy to the original full-precision LLM models, using the absolute max scaling. Subsequently, all data are compressed to 4-bit using our proposed significant data razoring (SDR) technique, which retains only the four most salient bits while discarding the others. Furthermore, we present an integer-based arithmetic unit dedicated to QRazor, enabling direct low-precision arithmetic operations without decompressing the SDR data. Despite the reduced quantization effort, QRazor achieves LLM accuracies better or comparable to state-of-the-art 4-bit methods. By also validating the hardware efficiency, our decompression-free arithmetic unit achieves 61.2\% and 57.8\% reduction in area and power consumption, respectively.
Supplementary Material: zip
Primary Area: infrastructure, software libraries, hardware, systems, etc.
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 9047
Loading