paper-with-me

홈 › Papers

CodeGEMM: A Codebook-Centric Approach to Efficient GEMM in Quantized LLMs

2025-12-19 · Gunho Park, Jeongin Bae, Byeongwook Kim, Baeseong park, Jiwon Ryu, Hoseung Kim, Se Jung Kwon, Dongsoo Lee arxiv

Weight-only quantization is widely used to mitigate the memory-bound nature of LLM inference. Codebook-based methods extend this trend by achieving strong accuracy in the extremely low-bit regime (e.g., 2-bit). However, current kernels rely on dequantization, which repeatedly fetches centroids and reconstructs weights, incurring substantial latency and cache pressure. We present CodeGEMM, a codebook-centric GEMM kernel that replaces dequantization with precomputed inner products between centroids and activations stored in a lightweight Psumbook. At inference, code indices directly gather these partial sums, eliminating per-element lookups and reducing the on-chip footprint. The kernel supports the systematic exploration of latency-memory-accuracy trade-offs under a unified implementation. On Llama-3 models, CodeGEMM delivers 1.83x (8B) and 8.93x (70B) speedups in the 2-bit configuration compared to state-of-the-art codebook-based quantization at comparable accuracy and further improves computing efficiency and memory subsystem utilization.

📄 PDF Abstract BibTeX arXiv:2512.17970

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

CodeGemma: Open Code Models Based on Gemma

2024-06-17 · CodeGemma Team, Heri Zhao, Jeffrey Hui, Joshua Howland 외

This paper introduces CodeGemma, a collection of specialized open code models built on top of Gemma, capable of a variety of code and natural language generation tasks. We release three model variants. CodeGemma 7B pretr…

Code CompletionMathematical ReasoningNatural Language UnderstandingText Generation

EVA: Accelerating LLM Decoding via an Efficient Vector Quantization Architecture

2026-05-22 · Bowen Duan, Cong Guo, Chiyue Wei, Haoxuan Shan 외 arxiv

Large Language Models (LLMs) have achieved impressive performance across diverse domains but remain inefficient during the autoregressive decoding phase. Unlike the prefill stage, which employs compute-bound GEMM operati…

DocuMint: Docstring Generation for Python using Small Language Models

2024-05-16 · Bibek Poudel, Adam Cook, Sekou Traore, Shelah Ameli

Effective communication, specifically through documentation, is the beating heart of collaboration among contributors in software development. Recent advancements in language models (LMs) have enabled the introduction of…

BenchmarkingCode Generation

Large Language Models in Code Co-generation for Safe Autonomous Vehicles

2025-05-26 · Ali Nouri, Beatriz Cabrero-Daniel, Zhennan Fei, Krishna Ronanki 외

Software engineers in various industrial domains are already using Large Language Models (LLMs) to accelerate the process of implementing parts of software systems. When considering its potential use for ADAS or AD syste…

Autonomous VehiclesCode Generation

Zero-Shot Attribution for Large Language Models: A Distribution Testing Approach

2025-06-25 · Clément L. Canonne, Yash Pote, Uddalok Sarkar

A growing fraction of all code is sampled from Large Language Models (LLMs). We investigate the problem of attributing code generated by language models using hypothesis testing to leverage established techniques and gua…