CoLLIE: Continual Learning of Language Grounding from Language-Image Embeddings

TL;DR

CoLLIE is a continual learning model that adapts language embeddings in multimodal spaces like CLIP, enabling efficient learning of new language use with minimal interference, demonstrated on referring expression tasks.

Contribution

It introduces a transformation-based approach for continual language grounding in vision models, enhancing generalization and few-shot learning capabilities.

Findings

Effective adaptation to new language use with few examples

Minimal impact on original zero-shot performance

Successful application to referring expression tasks

Abstract

This paper presents CoLLIE: a simple, yet effective model for continual learning of how language is grounded in vision. Given a pre-trained multimodal embedding model, where language and images are projected in the same semantic space (in this case CLIP by OpenAI), CoLLIE learns a transformation function that adjusts the language embeddings when needed to accommodate new language use. This is done by predicting the difference vector that needs to be applied, as well as a scaling factor for this vector, so that the adjustment is only applied when needed. Unlike traditional few-shot learning, the model does not just learn new classes and labels, but can also generalize to similar language use and leverage semantic compositionality. We verify the model's performance on two different tasks of identifying the targets of referring expressions, where it has to learn new language use. The results show that the model can efficiently learn and generalize from only a few examples, with little interference with the model's original zero-shot performance.