CLMay 20, 2022

Visually-Augmented Language Modeling

Microsoft
arXiv:2205.10178v222 citationsh-index: 102Has Code
Originality Highly original
AI Analysis

This addresses the limitation of text-only language models in handling visual knowledge-intensive tasks, offering a novel approach for multimodal AI applications.

The paper tackles the problem of language models lacking visual grounding by proposing VaLM, a pre-training framework that augments text with retrieved images, resulting in substantial gains on visual commonsense reasoning tasks.

Human language is grounded on multimodal knowledge including visual knowledge like colors, sizes, and shapes. However, current large-scale pre-trained language models rely on text-only self-supervised training with massive text data, which precludes them from utilizing relevant visual information when necessary. To address this, we propose a novel pre-training framework, named VaLM, to Visually-augment text tokens with retrieved relevant images for Language Modeling. Specifically, VaLM builds on a novel latent text-image alignment method via an image retrieval module to fetch corresponding images given a textual context. With the visually-augmented context, VaLM uses a visual knowledge fusion layer to enable multimodal grounded language modeling by attending to both text context and visual knowledge in images. We evaluate VaLM on various visual knowledge-intensive commonsense reasoning tasks, which require visual information to excel. The experimental results illustrate that VaLM outperforms all strong language-only and vision-language baselines with substantial gains in reasoning object commonsense including color, size, and shape. Our code is available at https://github.com/Victorwz/VaLM.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes