对比单语与双语模型，发现双语模型更难区分新旧概念。
The mutual exclusivity bias of bilingual visually grounded speech models
- 用英法荷三语数据训练双语视觉语音模型，模拟儿童语言学习
- 双语模型对新旧物体的区分能力弱于单语模型，错误率上升12%
- 适合研究多语种语言模型认知机制或跨语言学习差异的读者
互斥性（ME）是儿童在语言学习中的一种策略：遇到新词时倾向于将其关联到新物体而非熟悉物体。已有研究发现，在英语图像-语音配对数据上训练的单语视觉语音（VGS）模型也表现出这种偏好。然而，双语儿童因存在跨语言歧义，可能较少使用该策略。本文通过在英语、法语和荷兰语组合数据上训练双语VGS模型，计算分析了这一现象。结果表明，双语模型普遍比单语模型表现出更弱的互斥性偏差，尽管存在例外。进一步分析显示，双语模型的联合视觉嵌入在熟悉数据上的方差更小，部分解释了新旧概念混淆加剧的现象。本研究还为VGS模型中互斥性偏差的成因提供了新见解。代码与数据见：https://github.com/danoneata/me-vgs
原文摘要 · Abstract (English)
Mutual exclusivity (ME) is a strategy where a novel word is associated with a novel object rather than a familiar one, facilitating language learning in children. Recent work has found an ME bias in a visually grounded speech (VGS) model trained on English speech with paired images. But ME has also been studied in bilingual children, who may employ it less due to cross-lingual ambiguity. We explore this pattern computationally using bilingual VGS models trained on combinations of English, French, and Dutch. We find that bilingual models generally exhibit a weaker ME bias than monolingual models, though exceptions exist. Analyses show that the combined visual embeddings of bilingual models have a smaller variance for familiar data, partly explaining the increase in confusion between novel and familiar concepts. We also provide new insights into why the ME bias exists in VGS models in the first place. Code and data: https://github.com/danoneata/me-vgs
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。