Forensic authorship profiling uses linguistic markers to infer
characteristics about an author of a text. This task is paralleled in dialect
classification, where a prediction is made about the linguistic variety of a
text based on the text itself. While there have been significant advances in
the last years in variety classification (Jauhiainen et al., 2019) and
state-of-the-art approaches reach accuracies of up to 100% depending on the
similarity of varieties and the scope of prediction (e.g., Milne et al., 2012;
Blodgett et al., 2017), forensic linguistics rarely relies on these approaches
due to their lack of transparency (see Nini, 2023), amongst other reasons. In
this paper we therefore explore explainability of machine learning approaches
considering the forensic context. We focus on variety classification as a means
of geolinguistic profiling of unknown texts. For this we work with an approach
proposed by Xie et al. (2024) to extract the lexical items most relevant to the
variety classifications. We find that the extracted lexical features are indeed
representative of their respective varieties and note that the trained models
also rely on place names for classifications.

在这篇论文中，我们探讨了在法医语言学背景下机器学习方法的可解释性，通过使用一种地理语言学分析方法，提取与语言分类相关的词汇特征，并观察到训练模型也使用地名进行分类。

机器学习方法在法医语言学中的可解释性：地域语言作者特征研究案例

Explainability of Machine Learning Approaches in Forensic Linguistics: A  Case Study in Geolinguistic Authorship Profiling

Dialect classification is used in a variety of applications, such as machine
translation and speech recognition, to improve the overall performance of the
system. In a real-world scenario, a deployed dialect classification model can
encounter anomalous inputs that differ from the training data distribution,
also called out-of-distribution (OOD) samples. Those OOD samples can lead to
unexpected outputs, as dialects of those samples are unseen during model
training. Out-of-distribution detection is a new research area that has
received little attention in the context of dialect classification. Towards
this, we proposed a simple yet effective unsupervised Mahalanobis distance
feature-based method to detect out-of-distribution samples. We utilize the
latent embeddings from all intermediate layers of a wav2vec 2.0
transformer-based dialect classifier model for multi-task learning. Our
proposed approach outperforms other state-of-the-art OOD detection methods
significantly.

我们提出了一个简单而有效的无监督马氏距离特征方法，在流行的 wav2vec 2.0 变换器式方言分类器模型的所有中间层的潜在嵌入上进行多任务学习，用于检测超出分布的样本，并在 OOD 检测方法方面显著优于其他最先进的方法。

基于马氏距离的非监督式外部方言检测

Unsupervised Out-of-Distribution Dialect Detection with Mahalanobis  Distance

The pre-trained multi-lingual XLSR model generalizes well for language
identification after fine-tuning on unseen languages. However, the performance
significantly degrades when the languages are not very distinct from each
other, for example, in the case of dialects. Low resource dialect
classification remains a challenging problem to solve. We present a new data
augmentation method that leverages model training dynamics of individual data
points to improve sampling for latent mixup. The method works well in
low-resource settings where generalization is paramount. Our datamaps-based
mixup technique, which we call Map-Mix improves weighted F1 scores by 2%
compared to the random mixup baseline and results in a significantly
well-calibrated model. The code for our method is open sourced on
this https URL.

本文提出了一种名为 Map-Mix 的数据增广方法，用于提高低资源方言分类准确性，该方法通过数据点的模型训练动态来改进隐式 Mixup 的采样，相比于随机 Mixup 基线，Map-Mix 方法的加权 F1 得分提高了 2％，结果呈现出明显的良好校准模型。