Saved in:
Bibliographic Details
Main Authors: Na, Hongbin, Wang, Zimu, Maimaiti, Mieradilijiang, Chen, Tong, Wang, Wei, Shen, Tao, Chen, Ling
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2402.10699
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913970234851328
author Na, Hongbin
Wang, Zimu
Maimaiti, Mieradilijiang
Chen, Tong
Wang, Wei
Shen, Tao
Chen, Ling
author_facet Na, Hongbin
Wang, Zimu
Maimaiti, Mieradilijiang
Chen, Tong
Wang, Wei
Shen, Tao
Chen, Ling
contents Large language models (LLMs) have demonstrated promising potential in various downstream tasks, including machine translation. However, prior work on LLM-based machine translation has mainly focused on better utilizing training data, demonstrations, or pre-defined and universal knowledge to improve performance, with a lack of consideration of decision-making like human translators. In this paper, we incorporate Thinker with the Drift-Diffusion Model (Thinker-DDM) to address this issue. We then redefine the Drift-Diffusion process to emulate human translators' dynamic decision-making under constrained resources. We conduct extensive experiments under the high-resource, low-resource, and commonsense translation settings using the WMT22 and CommonMT datasets, in which Thinker-DDM outperforms baselines in the first two scenarios. We also perform additional analysis and evaluation on commonsense translation to illustrate the high effectiveness and efficacy of the proposed method.
format Preprint
id arxiv_https___arxiv_org_abs_2402_10699
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Thinker-DDM: Modeling Deliberation for Machine Translation with a Drift-Diffusion Process
Na, Hongbin
Wang, Zimu
Maimaiti, Mieradilijiang
Chen, Tong
Wang, Wei
Shen, Tao
Chen, Ling
Computation and Language
Large language models (LLMs) have demonstrated promising potential in various downstream tasks, including machine translation. However, prior work on LLM-based machine translation has mainly focused on better utilizing training data, demonstrations, or pre-defined and universal knowledge to improve performance, with a lack of consideration of decision-making like human translators. In this paper, we incorporate Thinker with the Drift-Diffusion Model (Thinker-DDM) to address this issue. We then redefine the Drift-Diffusion process to emulate human translators' dynamic decision-making under constrained resources. We conduct extensive experiments under the high-resource, low-resource, and commonsense translation settings using the WMT22 and CommonMT datasets, in which Thinker-DDM outperforms baselines in the first two scenarios. We also perform additional analysis and evaluation on commonsense translation to illustrate the high effectiveness and efficacy of the proposed method.
title Thinker-DDM: Modeling Deliberation for Machine Translation with a Drift-Diffusion Process
topic Computation and Language
url https://arxiv.org/abs/2402.10699