English(EN) When Can You Prune Your Network? A Study of Intermediate Neurons in Multilingual Speech Parsing

新的语音解析架构在不损失性能的情况下减少了参数

作者 PulseAugur 编辑部 · [1 个来源] · 2026-10-09 04:00

研究人员开发了一种更简单的端到端语音解析架构，消除了中间神经网络单元，将参数减少了 12%，同时保持或提高了性能。这种新方法表明，当预训练编码器被冻结时，中间神经网络单元主要有利于弥合表示差距。该研究对法语以及资源较少的斯洛文尼亚语和 Naija 语进行了全面评估，还考察了训练数据大小和预训练语音编码器层的影响。 AI

影响这项研究提供了一种更有效的语音解析方法，有可能降低计算成本并提高低资源语言的性能。

排序理由关于模型架构和评估的学术论文。[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: 15 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 关于模型架构和评估的学术论文。[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, infra; 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) · Minnie Kabra, Benjamin Lecouteux, Maximin Coavoux · 2026-10-09 04:00

何时可以修剪你的网络？一项关于多语言语音解析中中间神经元的研究

arXiv:2610.11520v1 Announce Type: new Abstract: End-to-end speech parsing, a task recently proposed, consists in predicting both the transcription and the syntactic tree for a spoken utterance. Existing architectures for speech parsing often utilise intermediate neural networks. …

本文如何被排名

报道来源 [1]

何时可以修剪你的网络？一项关于多语言语音解析中中间神经元的研究

相关实体

相关话题