Scene Graph Modification Based on Natural Language Commands

TL;DR

This paper introduces a novel approach for updating scene graphs based on natural language commands, leveraging graph-based transformers and cross attention, and provides datasets for future research.

Contribution

It presents new models for scene graph modification using transformers and releases datasets, addressing a gap in direct graph manipulation from natural language commands.

Findings

Proposed models outperform previous systems in scene graph modification tasks.

Introduced large datasets for scene graph editing based on natural language commands.

Demonstrated the effectiveness of graph-based sparse transformers and cross attention.

Abstract

Structured representations like graphs and parse trees play a crucial role in many Natural Language Processing systems. In recent years, the advancements in multi-turn user interfaces necessitate the need for controlling and updating these structured representations given new sources of information. Although there have been many efforts focusing on improving the performance of the parsers that map text to graphs or parse trees, very few have explored the problem of directly manipulating these representations. In this paper, we explore the novel problem of graph modification, where the systems need to learn how to update an existing scene graph given a new user's command. Our novel models based on graph-based sparse transformer and cross attention information fusion outperform previous systems adapted from the machine translation and graph generation literature. We further contribute our large graph modification datasets to the research community to encourage future research for this new problem.