Hybrid Batch Composition Effects on ALBEF CLIPScore in Zero-Shot Image-Text Retrieval

Information retrieval across different languages is an increasingly important challenge in natural language processing. Recent approaches based on multilingual pre-trained language models have achieved remarkable success, yet they often optimize for either monolingual, cross-lingual, or multilingual retrieval performance at the expense of others. This paper proposes a novel hybrid batch training strategy to simultaneously improve zero-shot retrieval performance across monolingual, cross-lingual, and multilingual settings while mitigating language bias. The approach fine-tunes multilingual lang

Research goal: How does adjusting the hybrid batch composition ratio (e.g., 30% monolingual, 30% cross-lingual, 40% multilingual) affect the CLIPScore performance of ALBEF models on zero-shot image-text retrieval tasks compared to uniform ratios?

Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.5/10.