LoRA微调中，顶1预测失效，改用梯度范数更准
When Top-1 Fails: Calibrating LoRA Monitors for Masked Diffusion LMs
- 用最大LoRA梯度范数替代顶1预测作为训练稳定性信号
- 在LLaDA族上实现0.68精度与0.79的F1值，优于传统方法
- 适用于短时程扩散语言模型的微调监控，不具普适性
离散扩散语言模型（DLM）微调继承了去噪时间置信度监控的低成本诊断机制，但其在参数高效微调（PEFT）中的意义尚未验证。本文测试顶1预测集中度作为崩溃预警信号，在三个DLM家族共816种LoRA/PEFT配置中，该预警对所有配置均触发，但日志显示200步时实际崩溃为0/816，精确率为零。原因在于预平衡饱和：顶1集中度在优化前已很高，迅速失去对最终训练稳定性的敏感性。随后评估最大LoRA梯度范数，这是一种基于参数的信号，反映梯度路由而非词元集中度。在合并的留出集（LLaDA族）上，经训练优化的阈值可识别出损失最低的十分之一配置，精度达0.68，F1为0.79，优于全正样本的顶1基线，甚至在较低的分组自助置信边界下仍表现更好。自回归控制和跨家族阈值失败表明结果仅适用于短时程DLM-LoRA检查，而非通用崩溃检测器。建议工作流：弃用顶1作为PEFT警报，早期记录最大梯度，按DLM家族校准阈值后再进行运行检查。
原文摘要 · Abstract (English)
Discrete diffusion language model (DLM) fine-tuning inherits inexpensive diagnostics from denoising-time confidence monitors, but their PEFT-training meaning is untested. We test top-1 argmax concentration as a collapse warning. Across 816 LoRA/PEFT configurations from three DLM families, the warning fires for every configuration while logs record 0/816 actual collapses at the 200 step horizon, giving zero precision. The cause is pre-equilibrium saturation: top-1 concentration is already high before optimization and quickly becomes insensitive to final training stability. We then evaluate max LoRA gradient norm, a parameter-side signal that samples gradient routing rather than token concentration. On a pooled held-out LLaDA-family split, a train-optimized threshold identifies top-decile final-loss configurations with precision 0.68 and F1=0.79, above the all-positive top-1 baseline even at the lower split-bootstrap confidence bound. Autoregressive controls and cross-family threshold failures bound the result to short-horizon DLM-LoRA inspection rather than a universal collapse detector. Workflow: drop top-1 as a PEFT alarm, log max-gradient early in training, and calibrate thresholds per DLM family before routing runs for inspection.
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。