English(EN) Pruning as Regularization: Sensitivity-Aware One-Shot Pruning in ASR

剪枝Whisper-small通过充当正则化器来提高ASR准确性

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-15 04:00

研究人员已经证明，神经网络剪枝可以作为自动语音识别（ASR）系统的正则化技术，而不仅仅是压缩方法。通过分析Whisper-small模型中不同组件的敏感性，他们发现剪枝特定层，如解码器自注意力或最后一个编码器层，实际上可以提高泛化能力并降低词错误率（WER）。这种方法允许更积极的压缩而不会显著损失准确性，表明剪枝可以成为一个有价值的架构设计工具。 AI

影响表明剪枝可用作改进模型泛化和压缩的设计工具，可能影响未来的ASR模型开发。

排序理由详细介绍模型剪枝新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 13 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 详细介绍模型剪枝新方法的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CL TIER_1 English(EN) · Julian Irigoyen, Arthur S\"ohler, Andreas S{\o}eborg Kirkedal · 2026-09-15 04:00

剪枝作为正则化：ASR中的感知敏感度单次剪枝

arXiv:2511.08092v2 Announce Type: replace-cross Abstract: We challenge the conventional view of neural network pruning as solely a compression technique, demonstrating that one-shot magnitude pruning serves as a powerful implicit regularizer for ASR. Using Whisper-small, we combi…

本文如何被排名

报道来源 [1]

剪枝作为正则化：ASR中的感知敏感度单次剪枝

相关实体

相关话题