Distilling Multi-Scale Knowledge for Event Temporal Relation Extraction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yao, Hao-Ren, Breitfeller, Luke, Naik, Aakanksha, Zhou, Chunxiao, Rose, Carolyn
Format: Preprint
Published: 2022
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909268357152768
author Yao, Hao-Ren
Breitfeller, Luke
Naik, Aakanksha
Zhou, Chunxiao
Rose, Carolyn
author_facet Yao, Hao-Ren
Breitfeller, Luke
Naik, Aakanksha
Zhou, Chunxiao
Rose, Carolyn
contents Event Temporal Relation Extraction (ETRE) is paramount but challenging. Within a discourse, event pairs are situated at different distances or the so-called proximity bands. The temporal ordering communicated about event pairs where at more remote (i.e., ``long'') or less remote (i.e., ``short'') proximity bands are encoded differently. SOTA models have tended to perform well on events situated at either short or long proximity bands, but not both. Nonetheless, real-world, natural texts contain all types of temporal event-pairs. In this paper, we present MulCo: Distilling Multi-Scale Knowledge via Contrastive Learning, a knowledge co-distillation approach that shares knowledge across multiple event pair proximity bands to improve performance on all types of temporal datasets. Our experimental results show that MulCo successfully integrates linguistic cues pertaining to temporal reasoning across both short and long proximity bands and achieves new state-of-the-art results on several ETRE benchmark datasets.
format Preprint
id arxiv_https___arxiv_org_abs_2209_00568
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle Distilling Multi-Scale Knowledge for Event Temporal Relation Extraction
Yao, Hao-Ren
Breitfeller, Luke
Naik, Aakanksha
Zhou, Chunxiao
Rose, Carolyn
Computation and Language
Artificial Intelligence
Machine Learning
Event Temporal Relation Extraction (ETRE) is paramount but challenging. Within a discourse, event pairs are situated at different distances or the so-called proximity bands. The temporal ordering communicated about event pairs where at more remote (i.e., ``long'') or less remote (i.e., ``short'') proximity bands are encoded differently. SOTA models have tended to perform well on events situated at either short or long proximity bands, but not both. Nonetheless, real-world, natural texts contain all types of temporal event-pairs. In this paper, we present MulCo: Distilling Multi-Scale Knowledge via Contrastive Learning, a knowledge co-distillation approach that shares knowledge across multiple event pair proximity bands to improve performance on all types of temporal datasets. Our experimental results show that MulCo successfully integrates linguistic cues pertaining to temporal reasoning across both short and long proximity bands and achieves new state-of-the-art results on several ETRE benchmark datasets.
title Distilling Multi-Scale Knowledge for Event Temporal Relation Extraction
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2209.00568