提出新方法让低比特模型微调更快更准，适合部署
Fine-Tuning Low-Bit Models with Gradient in Quantized Code Space
- 用可部署代码空间的梯度信号加速优化
- 在多种量化类型上均提升微调效果
- 兼顾效率与部署一致性，适合实际落地
低比特模型微调旨在适应量化模型的同时，保持最终部署检查点为相同低比特形式，这对降低存储和推理成本至关重要。在此约束下，适应成为在量化码和尺度上的优化问题。现有连续低比特训练虽高效，但易受直通估计误差或后量化差距影响；离散搜索虽符合部署要求，但在有限训练预算下通常效率过低。本文提出代码代理梯度（GradCodes），作为可部署代码空间的一阶信号以加速优化，并采用引导搜索策略保持部署一致性。跨算术推理、指令遵循和结构化语言理解任务的实验表明，GradCodes在不同量化数据类型下均能一致提升低比特模型微调性能。代码已开源。
原文摘要 · Abstract (English)
Fine-tuning Low-bit models aims to adapt a quantized model while keeping the final deployed checkpoint in the same low-bit form. This setting is practically important as it reduces memory and inference cost for storage and deployment. Under this constraint, adaptation becomes an optimization problem over quantization codes and scales. Existing continuous low-bit training is efficient, but it can be distorted by straight through estimation error or by post-quantize gap; discrete search is deployment-faithful, but it is often too inefficient under a finite training budget. We propose code surrogate gradient as the first order signal in deployable code space to acceleate optimization, and performing guided search to preserve deployment faithfulness. Experiments across arithmetic reasoning, instruction following, and structured language understanding show that GradCodes consistently improves fine-tuning low-bit models across different quantization datatypes. Code is provided at https://github.com/ovo67/GradCodes.
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。