How Far Do Foundation Models Transfer to Infant Signals? A Cross-Dataset Transfer Audit with a Unified Need Ontology
作者: Wu Hangyu
分类: cs.CL, cs.AI
发布日期: 2026-08-10
备注: 18 pages, 7 figures. Under review at AAAI 2027
💡 一句话要点
提出跨数据集转移审计方法以解决婴儿哭声信号分析问题
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 婴儿哭声分析 跨数据集转移 多层次审计 联合训练 领域自适应预训练
📋 核心要点
- 现有婴儿哭声分析方法主要依赖单一语料库,导致评估结果不全面,且标签不兼容问题严重。
- 论文提出了一种跨数据集转移审计的方法,通过多层次审计技术来评估不同语料库间的转移能力。
- 实验结果显示,跨语料库转移效果普遍为负,但在特定条件下,处理噪声语料库的效果显著提升,且联合训练表现优于简单标签合并。
📝 摘要(中文)
公共婴儿哭声语料库通常较小且标签不兼容,且评估通常只针对单一语料库。本文通过多层次泄漏审计,探讨了四个哭声语料库的转移能力,揭示了单一语料库评估的隐含问题。研究表明,跨语料库转移平均为负,且存在标签不一致的问题。尽管如此,研究提供了一种有效的处理小型噪声语料库的实用方案,并发布了相应的本体、映射代码和审计流程。
🔬 方法详解
问题定义:本文旨在解决婴儿哭声信号分析中,因语料库小、标签不兼容而导致的评估不准确问题。现有方法往往只针对单一语料库进行评估,忽视了跨语料库的转移能力。
核心思路:通过多层次泄漏审计,分析不同语料库间的转移效果,并提出在处理小型噪声语料库时的有效策略,旨在提高模型的泛化能力和实用性。
技术框架:研究采用了统一的五类需求本体,结合四个冻结编码器和手工基线进行评估。审计流程包括字节级和嵌入级去重,以及语料库内训练-测试近重复审计。
关键创新:论文的创新在于提出了跨数据集的转移审计方法,揭示了单一语料库评估的局限性,并提供了有效的处理小型噪声语料库的方案。与现有方法相比,强调了联合训练的优势。
关键设计:在实验中,采用了稳定的微调和领域自适应预训练策略,评估了不同标签预算下的模型表现,特别是在5-10次样本的情况下,领域自适应预训练表现优于稳定微调。
🖼️ 关键图片
📊 实验亮点
实验结果显示,跨语料库转移的平均效果为负,负转移比例在0.19-0.35之间,且在30个方向性单元中有18个显著。然而,在去除近重复后,处理噪声语料库的转移效果始终为正,联合训练在所有四个编码器-目标组合中表现优异,提升幅度可达37 F1点。
🎯 应用场景
该研究的潜在应用领域包括婴儿哭声分析、早期儿童发展监测及相关医疗领域。通过提供可用的联合训练资源,研究为处理小型、噪声数据集提供了新的思路,具有重要的实际价值和未来影响。
📄 摘要(原文)
Public infant cry corpora are small, label-incompatible, and almost always evaluated one corpus at a time. We ask what this practice hides and what fixes it. Across four cry corpora screened by a multi-level leakage audit (byte-level and embedding-level deduplication plus a within-corpus train-test near-duplicate audit), we probe four frozen encoders and a handcrafted baseline under a unified five-class need ontology and shared task formulations. The audit exposes what single-corpus evaluation conceals: within-domain macro-F1 swings by 0.57-0.80 for the same encoder, cross-corpus transfer is negative on average (negative-transfer ratio 0.19-0.35, significant in 18 of 30 directed cells, BH-FDR), and 349 content-identical clip groups carry conflicting metadata labels across corpus distributions. The same audit, however, reveals a consistent way forward. Transfer into the noisiest corpus is consistently positive in effect size at matched training size and after near-duplicate removal, offering a practical recipe for small, noisy corpora. Frozen probes saturate at modest label budgets, while stabilized fine-tuning wins with full labels; domain-adaptive pretraining significantly beats stabilized fine-tuning at 5-10-shot (the 1-shot advantage is not robust to optimization-seed variance) but shows no significant advantage at 50-shot or beyond. In the tested binary, shared-label settings, ontology-mapped joint training wins in all four encoder-by-target combinations, whereas naively merging unmapped labels costs up to 37 F1 points. We release the ontology, mapping code, and audit pipeline, turning incompatible cry corpora into a usable joint-training resource.