扩散语言模型在量化后仍保持高编码性能，更适合低资源部署。
On the Quantization Robustness of Diffusion Language Models in Coding Benchmarks
- 采用扩散机制的编码模型在2-4位量化下表现更稳定。
- 相比自回归模型，其在HumanEval和MBPP上精度下降更小。
- 混合精度配置实现性能、速度与内存的平衡，适合实际部署。
自回归大语言模型在编码任务中表现优异，但内存与推理成本高昂。基于扩散的语言模型（d-LLMs）通过迭代去噪实现固定推理开销，但其在训练后量化（PTQ）下的行为研究较少。本文评估了GPTQ与改进的赫森感知量化（HAWQ）在扩散编码模型CoDA上的应用效果，发现其在2-4位低比特量化下相较于自回归模型Qwen3-1.7B具有更强鲁棒性，在HumanEval与MBPP基准上均表现出更小的精度损失。此外，由HAWQ生成的混合精度配置可在准确率、延迟与内存间实现平滑权衡。结果表明，扩散语言模型因更高的量化鲁棒性，可能更适合高效部署。
原文摘要 · Abstract (English)
Auto-regressive Large Language Models (LLMs) achieve strong performance on coding tasks, but incur high memory and inference costs. Diffusion-based language models (d-LLMs) offer bounded inference cost via iterative denoising, but their behavior under post-training quantization (PTQ) has been sparsely explored. We investigate the application and robustness of PTQ techniques, specifically GPTQ and a modified Hessian-Aware Quantization (HAWQ) algorithm, on a diffusion-based coding LLM (CoDA) and observe that these methods applied to CoDA exhibit greater robustness at low bitwidths compared to Qwen3-1.7B, its auto-regressive counterpart, under a standardized evaluation pipeline. We find that in our setup, CoDA exhibits greater robustness at low bitwidths (2-4 bits), with smaller accuracy degradation across HumanEval and MBPP benchmarks. Additionally, mixed-precision configurations derived from HAWQ provide smooth trade-offs across accuracy, latency, and memory. The results suggest that diffusion LLMs may offer advantages for efficient deployment due to more quantization-resilience.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。