Scalability of OWL-ViT's Zero-Shot Cross-Lingual Retrieval Accuracy Compared to CLIP and BLIP

Benefiting from transformer-based pre-trained language models, neural ranking models have made significant progress. More recently, the advent of multilingual pre-trained language models provides great support for designing neural cross-lingual retrieval models. However, due to unbalanced pre-training data in different languages, multilingual language models have already shown a performance gap between high and low-resource languages in many downstream tasks. And cross-lingual retrieval models built on such pre-trained models can inherit language bias, leading to suboptimal result for low-reso

Research goal: How does the scalability of OWL-ViT's zero-shot cross-lingual retrieval accuracy compare to CLIP and BLIP when trained on different intermediate English tasks and then evaluated on low-resource languages in XTREME-R, measured by model size and performance trade-offs?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 9.2/10.