Saved in:
Bibliographic Details
Main Authors: Betala, Siddharth, Raj, Kushan, Betala, Vipul, Saswade, Rohan
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2511.07010
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918193711284224
author Betala, Siddharth
Raj, Kushan
Betala, Vipul
Saswade, Rohan
author_facet Betala, Siddharth
Raj, Kushan
Betala, Vipul
Saswade, Rohan
contents 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. The judge component classifies translations into three categories: correct, visually ambiguous (requiring image context), or mistranslated (poor translation quality). Identified errors are routed to specialized correctors: GPT-4o-mini regenerates captions requiring visual disambiguation, while IndicTrans2 retranslates cases with pure translation quality issues. This automated pipeline processes 28,928 training examples across four languages, correcting an average of 17.1% of captions per language. We then apply Low-Rank Adaptation (LoRA) to fine-tune the IndicTrans2 en-indic 200M distilled model on both original and corrected datasets. Training on corrected data yields consistent improvements, with BLEU score gains of +1.30 for English-Bengali on the evaluation set (42.00 -> 43.30) and +0.70 on the challenge set (44.90 -> 45.60), +0.60 for English-Odia on the evaluation set (41.00 -> 41.60), and +0.10 for English-Hindi on the challenge set (53.90 -> 54.00).
format Preprint
id arxiv_https___arxiv_org_abs_2511_07010
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Picture is Worth a Thousand (Correct) Captions: A Vision-Guided Judge-Corrector System for Multimodal Machine Translation
Betala, Siddharth
Raj, Kushan
Betala, Vipul
Saswade, Rohan
Computation and Language
Computer Vision and Pattern Recognition
Human-Computer Interaction
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. The judge component classifies translations into three categories: correct, visually ambiguous (requiring image context), or mistranslated (poor translation quality). Identified errors are routed to specialized correctors: GPT-4o-mini regenerates captions requiring visual disambiguation, while IndicTrans2 retranslates cases with pure translation quality issues. This automated pipeline processes 28,928 training examples across four languages, correcting an average of 17.1% of captions per language. We then apply Low-Rank Adaptation (LoRA) to fine-tune the IndicTrans2 en-indic 200M distilled model on both original and corrected datasets. Training on corrected data yields consistent improvements, with BLEU score gains of +1.30 for English-Bengali on the evaluation set (42.00 -> 43.30) and +0.70 on the challenge set (44.90 -> 45.60), +0.60 for English-Odia on the evaluation set (41.00 -> 41.60), and +0.10 for English-Hindi on the challenge set (53.90 -> 54.00).
title A Picture is Worth a Thousand (Correct) Captions: A Vision-Guided Judge-Corrector System for Multimodal Machine Translation
topic Computation and Language
Computer Vision and Pattern Recognition
Human-Computer Interaction
url https://arxiv.org/abs/2511.07010