On the Interplay between Positional Encodings, Morphological Complexity, and Word Order Flexibility

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Tatariya, Kushal, Poelman, Wessel, de Lhoneux, Miryam
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917073136910336
author Tatariya, Kushal
Poelman, Wessel
de Lhoneux, Miryam
author_facet Tatariya, Kushal
Poelman, Wessel
de Lhoneux, Miryam
contents Language model architectures are predominantly first created for English and subsequently applied to other languages. It is an open question whether this architectural bias leads to degraded performance for languages that are structurally different from English. We examine one specific architectural choice: positional encodings, through the lens of the trade-off hypothesis: the supposed interplay between morphological complexity and word order flexibility. This hypothesis posits a trade-off between the two: a more morphologically complex language can have a more flexible word order, and vice-versa. Positional encodings are a direct target to investigate the implications of this hypothesis in relation to language modelling. We pretrain monolingual model variants with absolute, relative, and no positional encodings for seven typologically diverse languages and evaluate them on four downstream tasks. Contrary to previous findings, we do not observe a clear interaction between position encodings and morphological complexity or word order flexibility, as measured by various proxies. Our results show that the choice of tasks, languages, and metrics are essential for drawing stable conclusions
format Preprint
id arxiv_https___arxiv_org_abs_2511_08139
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle On the Interplay between Positional Encodings, Morphological Complexity, and Word Order Flexibility
Tatariya, Kushal
Poelman, Wessel
de Lhoneux, Miryam
Computation and Language
Language model architectures are predominantly first created for English and subsequently applied to other languages. It is an open question whether this architectural bias leads to degraded performance for languages that are structurally different from English. We examine one specific architectural choice: positional encodings, through the lens of the trade-off hypothesis: the supposed interplay between morphological complexity and word order flexibility. This hypothesis posits a trade-off between the two: a more morphologically complex language can have a more flexible word order, and vice-versa. Positional encodings are a direct target to investigate the implications of this hypothesis in relation to language modelling. We pretrain monolingual model variants with absolute, relative, and no positional encodings for seven typologically diverse languages and evaluate them on four downstream tasks. Contrary to previous findings, we do not observe a clear interaction between position encodings and morphological complexity or word order flexibility, as measured by various proxies. Our results show that the choice of tasks, languages, and metrics are essential for drawing stable conclusions
title On the Interplay between Positional Encodings, Morphological Complexity, and Word Order Flexibility
topic Computation and Language
url https://arxiv.org/abs/2511.08139