Malta

Closely related languages show linguistic similarities that allow speakers of one language to understand speakers of another language without having actively learned it. Mutual intelligibility varies in degree and is typically tested in psycholinguistic experiments. To study mutual intelligibility computationally, we propose a computer-assisted method using the Linear Discriminative Learner, a computational model developed to approximate the cognitive processes by which humans learn languages, which we expand with multilingual semantic vectors and multilingual sound classes. We test the model on German, Dutch, and English, three closely related Germanic languages. We find that our model’s comprehension accuracy depends on 1) the automatic trimming of inflections and 2)the language pair for which comprehension is tested. Our multilingual modelling approach does not only offer new methodological findings for automatic testing of mutual intelligibility across languages but also extends the use of Linear Discriminative Learning to multilingual settings.

EACL 2024

A Computational Model for the Assessment of Mutual Intelligibility Among Closely Related Languages

germanic

computational linguistics

mutual intelligibility

workshop paper

Welcome to the 18th Conference of the European Chapter of the Association for Computational Linguistics (EACL). Continuing its mission of expanding and involving the science community of all European countries, EACL has selected the Malta community for the 18th EACL. Considering the importance of physical interaction among researchers, the conference will be held at the Hotel Radisson Blu, St. Julians, in Malta, from 17 to 22 of March, 2024. The conference will also feature the possibility to participate virtually. As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language. 
#### [Conference Handbook](https://drive.google.com/file/d/13Wn38q0yev6U3RTgnf4Tn5T9GuAuZ-xW/view?usp=sharing) *downloadable*

To access the EACL 2024 event page you need to register [**here**](https://acl.swoogo.com/EACL2024). 

Please register!

As the flagship European conference in the field of computational linguistics, EACL welcomes European and international researchers covering a broad spectrum of research areas that are concerned with computational approaches to natural language.

Comparing eye-movement data of the readers of different languages or language varieties enables us to delve into the linkage between language variation and cognitive processing pattern. Within this scope, the Chinese language is particularly attractive in terms of the uniqueness of its written system, and the diversity of written varieties. As the native language of a wide spectrum of Chinese language speakers, Mandarin contrasts with Cantonese essentially on the complexity and semantic transparency of a single character, and also considerably on word length variability and word order flexibility. These manifold variations back a hypothesis that Simplified Mandarin readers exert more cognitive effort on the context that the Traditional Cantonese readers. The computational language models (LM) accurately quantify the predictive processing (e.g. Surprisal) and contextual effect (e.g. Entropy and Entropy Reduction) underlying the real-time language processing, which allows a deeper and reproducible inspection of high-level factors and their manifestation on the variation of language processing over different written scripts. Leveraging both masked and incremental language models, this study focus on the feature importance of different LM features (Surprisal, Entropy, and Entropy Reduction) in the modeling of eye-movement behaviors (durations and saccades in both early and late stages). The results suggest that the variation of eye-movement patterns is more salient on newly introduced LM features while vague on the widely studied properties. Specifically, Entropy effect is larger in Mandarin reading while Surprisal effect is more pronounced in Cantonese reading.

Predicting Mandarin and Cantonese Adult Speakers’ Eye-Movement Patterns in Natural Reading

LAJaR (Language Atlas of Japanese and Ryukyuan) is a linguistic typology database focusing on micro-variation of the Japonic (Japanese and Ryukyuan) languages. This paper aims to report the design and progress of this ongoing database project. Finally, we also show a case study utilizing its database on zero copulas among the Japonic languages.

Language Atlas of Japanese and Ryukyuan (LAJaR): A Linguistic Typology Database for Endangered Japonic Languages

This paper lays the groundwork for initiating research into Source Language Identification; the task of identifying the original language of a machine-translated text. We contribute a dataset of translations from a typologically diverse spectrum of languages into English and use it to set initial baselines for this novel task.

GTNC: A Many-To-One Dataset of Google Translations from NewsCrawl

Emotion classification is a challenging task in NLP due to the inherent idiosyncratic and subjective nature of linguistic expression. The problem of processing multiple languages at once, each with their own variations of emotional expression, makes this task even more challenging for code-mixed data. Pre-trained language models (PLMs) have achieved high performance for many tasks and languages. It remains to be seen, however, whether these models learn and are robust to the differences in emotional expression across languages. Sociolinguistic studies have shown that Hinglish speakers make associations between language choice and emotion. Namely, they prefer Hindi when expressing negative emotions and English when expressing positive emotions. To understand if language models can pick up on these associations, we study the effect of language on emotion prediction across 3 PLMs on a Hinglish emotion classification dataset. Using LIME and token level language ID, we find that models do learn these associations between language choice and emotion. Moreover, having code-mixed data present in the pre-training can augment that learning even when task-specific data is scarce.

Sociolinguistically Informed Interpretability: A Case Study on Hinglish Emotion Classification

In order to draw generalizable conclusions about the performance of multilingual models across languages, it is important to evaluate on a set of languages that captures linguistic diversity. Linguistic typology is increasingly used to justify language selection, inspired by language sampling in linguistics. However, justifications for 'typological diversity' exhibit great variation, as there seems to be no set definition, methodology or consistent link to linguistic typology. In this work, we provide a systematic insight into how previous work in the ACL Anthology uses the term 'typological diversity'. Our two main findings are: 1) what is meant by typologically diverse language selection is not consistent and 2) the actual typological diversity of the language sets in these papers varies greatly. We argue that, when making claims about 'typological diversity', an operationalization of this should be included. A systematic approach that quantifies this claim, also with respect to the number of languages used, would be even better.

A Call for Consistency in Reporting Typological Diversity

In traditional studies on language evolution, scholars often emphasize the importance of sound laws and sound correspondences for phylogenetic inference of language family trees. However, to date, computational approaches have typically not taken this potential into account. Most computational studies still rely on lexical cognates as major data source for phylogenetic reconstruction in linguistics, although there do exist a few studies in which authors praise the benefits of comparing words at the level of sound sequences. Building on (a) ten diverse datasets from different language families, and (b) state-of-the-art methods for automated cognate and sound correspondence detection, we test, for the first time, the performance of sound-based versus cognate-based approaches to phylogenetic reconstruction. Our results show that phylogenies reconstructed from lexical cognates are topologically closer, by approximately one third with respect to the generalized quartet distance on average, to the gold standard phylogenies than phylogenies reconstructed from sound correspondences.

Are Sounds Sound for Phylogenetic Reconstruction?

In Universal Dependencies, compounds, which we understand as words containing two or more roots, are represented according to tokenization, which reflects the orthographic conventions of the language. A closed compound (e.g. water fall) corresponds to a single word in Universal Dependencies while a hyphenated compound (father-in-law) and an open compound (apple pie) to multiple words. The aim of this paper is to open a discussion on how to move towards a more consistent annotation of compounds. The solution we argue for is to represent the internal structure of all compound types analogously to syntactic phrases, which would not only increase the comparability of compounding within and across languages, but also allow comparisons of compounds and syntactic phrases.

Compounds in Universal Dependencies: A Survey in Five European Languages

While massively multilingual speech models like wav2vec 2.0 XLSR-128 can be directly fine-tuned for automatic speech recognition (ASR), downstream performance can still be relatively poor on languages that are under-represented in the pre-training data. Continued pre-training on 70–200 hours of untranscribed speech in these languages can help — but what about languages without that much recorded data? For such cases, we show that supplementing the target language with data from a similar, higher-resource 'donor' language can help. For example, continued pretraining on only 10 hours of low-resource Punjabi supplemented with 60 hours of donor Hindi is almost as good as continued pretraining on 70 hours of Punjabi. By contrast, sourcing supplemental data from less similar donors like Bengali does not improve ASR performance. To inform donor language selection, we propose a novel similarity metric based on the sequence distribution of induced acoustic units: the Acoustic Token Distribution Similarity (ATDS). Across a set of typologically different target languages (Punjabi, Galician, Iban, Setswana), we show that the ATDS between the target language and its candidate donors precisely predicts target language ASR performance.

Predicting positive transfer for improved low-resource speech recognition using acoustic pseudo-tokens

Large language models (LLMs) perform well on (at least) some evaluations of both few-shot multilingual adaptation and reasoning. However, evaluating the intersection of these two skills—multilingual few-shot reasoning—is difficult: even relatively low-resource languages can be found in large training corpora, raising the concern that when we intend to evaluate a model’s ability to generalize to a new language, that language may have in fact been present during the model’s training. If such language contamination has occurred, apparent cases of few-shot reasoning could actually be due to memorization. Towards understanding the capability of models to perform multilingual few-shot reasoning, we propose modeLing, a benchmark of Rosetta stone puzzles. This type of puzzle, originating from competitions called Linguistics Olympiads, contain a small number of sentences in a target language not previously known to the solver. Each sentence is translated to the solver’s language such that the provided sentence pairs uniquely specify a single most reasonable underlying set of rules; solving requires applying these rules to translate new expressions (Figure 1). modeLing languages are chosen to be extremely low-resource such that the risk of training data contamination is low, and unlike prior datasets, it consists entirely of problems written specifically for this work, as a further measure against data leakage. Empirically, we find evidence that popular LLMs do not have data leakage on our benchmark.

ModeLing: A Novel Dataset for Testing Linguistic Reasoning in Language Models

We present our submission to the unconstrained subtask of the SIGTYP 2024 Shared Task on Word Embedding Evaluation for Ancient and Historical Languages for morphological annotation, POS-tagging, lemmatization, characterand word-level gap-filling. We developed a simple, uniform, and computationally lightweight approach based on the adapters framework using parameter-efficient fine-tuning. We applied the same adapter-based approach uniformly to all tasks and 16 languages by fine-tuning stacked language- and task-specific adapters. Our submission obtained an overall second place out of three submissions, with the first place in word-level gap-filling. Our results show the feasibility of adapting language models pre-trained on modern languages to historical and ancient languages via adapter training.

Premium content

A Computational Model for the Assessment of Mutual Intelligibility Among Closely Related Languages

Downloads

Next from EACL 2024

Predicting Mandarin and Cantonese Adult Speakers’ Eye-Movement Patterns in Natural Reading

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES