Can Graph Neural Networks Learn Language with Extremely Weak Text Supervision?

TL;DR

This paper introduces a multi-modal prompt learning approach that enables Graph Neural Networks to perform well with minimal and weak text supervision, achieving zero-shot classification and strong transferability across tasks and domains.

Contribution

It proposes a novel paradigm that embeds graphs in the same space as LLMs by learning graph and text prompts simultaneously, addressing data scarcity and domain gaps.

Findings

Superior performance in few-shot and multi-task settings

Effective zero-shot classification with weak supervision

Generalization to unseen classes

Abstract

While great success has been achieved in building vision models with Contrastive Language-Image Pre-training (CLIP) over internet-scale image-text pairs, building transferable Graph Neural Networks (GNNs) with CLIP pipeline is challenging because of the scarcity of labeled data and text supervision, different levels of downstream tasks, and the conceptual gaps between domains. In this work, to address these issues, we propose a multi-modal prompt learning paradigm to effectively adapt pre-trained GNN to downstream tasks and data, given only a few semantically labeled samples, each with extremely weak text supervision. Our new paradigm embeds the graphs directly in the same space as the Large Language Models (LLMs) by learning both graph prompts and text prompts simultaneously. We demonstrate the superior performance of our paradigm in few-shot, multi-task-level, and cross-domain settings. Moreover, we build the first CLIP-style zero-shot classification prototype that can generalize GNNs to unseen classes with extremely weak text supervision. The code is available at https://github.com/Violet24K/Morpher.