World-to-Words: Grounded Open Vocabulary Acquisition through Fast Mapping in Vision-Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ma, Ziqiao, Pan, Jiayi, Chai, Joyce
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916542360322048
author Ma, Ziqiao
Pan, Jiayi
Chai, Joyce
author_facet Ma, Ziqiao
Pan, Jiayi
Chai, Joyce
contents The ability to connect language units to their referents in the physical world, referred to as grounding, is crucial to learning and understanding grounded meanings of words. While humans demonstrate fast mapping in new word learning, it remains unclear whether modern vision-language models can truly represent language with their grounded meanings and how grounding may further bootstrap new word learning. To this end, we introduce Grounded Open Vocabulary Acquisition (GOVA) to examine grounding and bootstrapping in open-world language learning. As an initial attempt, we propose object-oriented BERT (OctoBERT), a novel visually-grounded language model by pre-training on image-text pairs highlighting grounding as an objective. Through extensive experiments and analysis, we demonstrate that OctoBERT is a more coherent and fast grounded word learner, and that the grounding ability acquired during pre-training helps the model to learn unseen words more rapidly and robustly. Our code is available at https://github.com/sled-group/world-to-words
format Preprint
id arxiv_https___arxiv_org_abs_2306_08685
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle World-to-Words: Grounded Open Vocabulary Acquisition through Fast Mapping in Vision-Language Models
Ma, Ziqiao
Pan, Jiayi
Chai, Joyce
Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
The ability to connect language units to their referents in the physical world, referred to as grounding, is crucial to learning and understanding grounded meanings of words. While humans demonstrate fast mapping in new word learning, it remains unclear whether modern vision-language models can truly represent language with their grounded meanings and how grounding may further bootstrap new word learning. To this end, we introduce Grounded Open Vocabulary Acquisition (GOVA) to examine grounding and bootstrapping in open-world language learning. As an initial attempt, we propose object-oriented BERT (OctoBERT), a novel visually-grounded language model by pre-training on image-text pairs highlighting grounding as an objective. Through extensive experiments and analysis, we demonstrate that OctoBERT is a more coherent and fast grounded word learner, and that the grounding ability acquired during pre-training helps the model to learn unseen words more rapidly and robustly. Our code is available at https://github.com/sled-group/world-to-words
title World-to-Words: Grounded Open Vocabulary Acquisition through Fast Mapping in Vision-Language Models
topic Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2306.08685