Multimodal Contrastive In-Context Learning

TL;DR

This paper introduces a multimodal contrastive in-context learning framework that improves understanding, interpretability, and performance of large language models in multimodal tasks, especially under resource constraints.

Contribution

It presents a novel contrastive interpretation of ICL, addresses biases in multimodal input formatting, and proposes an on-the-fly ICL method for better performance in challenging scenarios.

Findings

Enhanced ICL performance on multimodal datasets.

Effective detection of hateful memes.

Improved interpretability of ICL mechanisms.

Abstract

The rapid growth of Large Language Models (LLMs) usage has highlighted the importance of gradient-free in-context learning (ICL). However, interpreting their inner workings remains challenging. This paper introduces a novel multimodal contrastive in-context learning framework to enhance our understanding of ICL in LLMs. First, we present a contrastive learning-based interpretation of ICL in real-world settings, marking the distance of the key-value representation as the differentiator in ICL. Second, we develop an analytical framework to address biases in multimodal input formatting for real-world datasets. We demonstrate the effectiveness of ICL examples where baseline performance is poor, even when they are represented in unseen formats. Lastly, we propose an on-the-fly approach for ICL (Anchored-by-Text ICL) that demonstrates effectiveness in detecting hateful memes, a task where typical ICL struggles due to resource limitations. Extensive experiments on multimodal datasets reveal that our approach significantly improves ICL performance across various scenarios, such as challenging tasks and resource-constrained environments. Moreover, it provides valuable insights into the mechanisms of in-context learning in LLMs. Our findings have important implications for developing more interpretable, efficient, and robust multimodal AI systems, especially in challenging tasks and resource-constrained environments.