A Picture is Worth a Thousand (Correct) Captions: A Vision-Guided Judge-Corrector System for Multimodal Machine Translation
Identifikátory výsledku
Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F26%3AXQBKGLZ6" target="_blank" >RIV/00216208:11320/26:XQBKGLZ6 - isvavai.cz</a>
Výsledek na webu
<a href="https://arxiv.org/abs/2511.07010" target="_blank" >https://arxiv.org/abs/2511.07010</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.48550/arXiv.2511.07010" target="_blank" >10.48550/arXiv.2511.07010</a>
Alternativní jazyky
Jazyk výsledku
angličtina
Název v původním jazyce
A Picture is Worth a Thousand (Correct) Captions: A Vision-Guided Judge-Corrector System for Multimodal Machine Translation
Popis výsledku v původním jazyce
In this paper, we describe our system under the team name BLEU Monday for the English-to-Indic Multimodal Translation Task at WAT 2025. We participate in the text-only translation tasks for English-Hindi, English-Bengali, English-Malayalam, and English-Odia language pairs. We present a two-stage approach that addresses quality issues in the training data through automated error detection and correction, followed by parameter-efficient model fine-tuning. Our methodology introduces a vision-augmented judge-corrector pipeline that leverages multimodal language models to systematically identify and correct translation errors in the training data.
Název v anglickém jazyce
A Picture is Worth a Thousand (Correct) Captions: A Vision-Guided Judge-Corrector System for Multimodal Machine Translation
Popis výsledku anglicky
In this paper, we describe our system under the team name BLEU Monday for the English-to-Indic Multimodal Translation Task at WAT 2025. We participate in the text-only translation tasks for English-Hindi, English-Bengali, English-Malayalam, and English-Odia language pairs. We present a two-stage approach that addresses quality issues in the training data through automated error detection and correction, followed by parameter-efficient model fine-tuning. Our methodology introduces a vision-augmented judge-corrector pipeline that leverages multimodal language models to systematically identify and correct translation errors in the training data.
Klasifikace
Druh
O - Ostatní výsledky
CEP obor
—
OECD FORD obor
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Návaznosti výsledku
Projekt
—
Návaznosti
—
Ostatní
Rok uplatnění
2025
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů