English

Towards Multimodal In-Context Learning for Vision & Language Models

Computer Vision and Pattern Recognition 2024-07-18 v2

Abstract

State-of-the-art Vision-Language Models (VLMs) ground the vision and the language modality primarily via projecting the vision tokens from the encoder to language-like tokens, which are directly fed to the Large Language Model (LLM) decoder. While these models have shown unprecedented performance in many downstream zero-shot tasks (eg image captioning, question answers, etc), still little emphasis has been put on transferring one of the core LLM capability of In-Context Learning (ICL). ICL is the ability of a model to reason about a downstream task with a few examples demonstrations embedded in the prompt. In this work, through extensive evaluations, we find that the state-of-the-art VLMs somewhat lack the ability to follow ICL instructions. In particular, we discover that even models that underwent large-scale mixed modality pre-training and were implicitly guided to make use of interleaved image and text information (intended to consume helpful context from multiple images) under-perform when prompted with few-shot demonstrations (in an ICL way), likely due to their lack of direct ICL instruction tuning. To enhance the ICL abilities of the present VLM, we propose a simple yet surprisingly effective multi-turn curriculum-based learning methodology with effective data mixes, leading up to a significant 21.03% (and 11.3% on average) ICL performance boost over the strongest VLM baselines and a variety of ICL benchmarks. Furthermore, we also contribute new benchmarks for ICL evaluation in VLMs and discuss their advantages over the prior art.

Keywords

Cite

@article{arxiv.2403.12736,
  title  = {Towards Multimodal In-Context Learning for Vision & Language Models},
  author = {Sivan Doveh and Shaked Perek and M. Jehanzeb Mirza and Wei Lin and Amit Alfassy and Assaf Arbelle and Shimon Ullman and Leonid Karlinsky},
  journal= {arXiv preprint arXiv:2403.12736},
  year   = {2024}
}
R2 v1 2026-06-28T15:25:45.281Z