Translation Deserves Better: Analyzing Translation Artifacts in Cross-lingual Visual Question Answering
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909216692764672 |
|---|---|
| author | Park, ChaeHun Lee, Koanho Lim, Hyesu Kim, Jaeseok Park, Junmo Heo, Yu-Jung Chang, Du-Seong Choo, Jaegul |
| author_facet | Park, ChaeHun Lee, Koanho Lim, Hyesu Kim, Jaeseok Park, Junmo Heo, Yu-Jung Chang, Du-Seong Choo, Jaegul |
| contents | Building a reliable visual question answering~(VQA) system across different languages is a challenging problem, primarily due to the lack of abundant samples for training. To address this challenge, recent studies have employed machine translation systems for the cross-lingual VQA task. This involves translating the evaluation samples into a source language (usually English) and using monolingual models (i.e., translate-test). However, our analysis reveals that translated texts contain unique characteristics distinct from human-written ones, referred to as translation artifacts. We find that these artifacts can significantly affect the models, confirmed by extensive experiments across diverse models, languages, and translation processes. In light of this, we present a simple data augmentation strategy that can alleviate the adverse impacts of translation artifacts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_02331 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Translation Deserves Better: Analyzing Translation Artifacts in Cross-lingual Visual Question Answering Park, ChaeHun Lee, Koanho Lim, Hyesu Kim, Jaeseok Park, Junmo Heo, Yu-Jung Chang, Du-Seong Choo, Jaegul Computation and Language Building a reliable visual question answering~(VQA) system across different languages is a challenging problem, primarily due to the lack of abundant samples for training. To address this challenge, recent studies have employed machine translation systems for the cross-lingual VQA task. This involves translating the evaluation samples into a source language (usually English) and using monolingual models (i.e., translate-test). However, our analysis reveals that translated texts contain unique characteristics distinct from human-written ones, referred to as translation artifacts. We find that these artifacts can significantly affect the models, confirmed by extensive experiments across diverse models, languages, and translation processes. In light of this, we present a simple data augmentation strategy that can alleviate the adverse impacts of translation artifacts. |
| title | Translation Deserves Better: Analyzing Translation Artifacts in Cross-lingual Visual Question Answering |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2406.02331 |