Saved in:
Bibliographic Details
Main Authors: Tseriotou, Talia, Tsakalidis, Adam, Liakata, Maria
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2408.15689
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916448471875584
author Tseriotou, Talia
Tsakalidis, Adam
Liakata, Maria
author_facet Tseriotou, Talia
Tsakalidis, Adam
Liakata, Maria
contents Dynamic representation learning plays a pivotal role in understanding the evolution of linguistic content over time. On this front both context and time dynamics as well as their interplay are of prime importance. Current approaches model context via pre-trained representations, which are typically temporally agnostic. Previous work on modelling context and temporal dynamics has used recurrent methods, which are slow and prone to overfitting. Here we introduce TempoFormer, the first task-agnostic transformer-based and temporally-aware model for dynamic representation learning. Our approach is jointly trained on inter and intra context dynamics and introduces a novel temporal variation of rotary positional embeddings. The architecture is flexible and can be used as the temporal representation foundation of other models or applied to different transformer-based architectures. We show new SOTA performance on three different real-time change detection tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2408_15689
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle TempoFormer: A Transformer for Temporally-aware Representations in Change Detection
Tseriotou, Talia
Tsakalidis, Adam
Liakata, Maria
Computation and Language
Dynamic representation learning plays a pivotal role in understanding the evolution of linguistic content over time. On this front both context and time dynamics as well as their interplay are of prime importance. Current approaches model context via pre-trained representations, which are typically temporally agnostic. Previous work on modelling context and temporal dynamics has used recurrent methods, which are slow and prone to overfitting. Here we introduce TempoFormer, the first task-agnostic transformer-based and temporally-aware model for dynamic representation learning. Our approach is jointly trained on inter and intra context dynamics and introduces a novel temporal variation of rotary positional embeddings. The architecture is flexible and can be used as the temporal representation foundation of other models or applied to different transformer-based architectures. We show new SOTA performance on three different real-time change detection tasks.
title TempoFormer: A Transformer for Temporally-aware Representations in Change Detection
topic Computation and Language
url https://arxiv.org/abs/2408.15689