Which One? Leveraging Context Between Objects and Multiple Views for Language Grounding

Anonymous

TL;DR: Real-world language use is situated in a 3D environment and must consider a rich context of alternatives. We design a model that explicitly considers both the context between objects and their views to improve language grounding.

Abstract: When connecting objects and their language referents in an embodied 3D environment, it is crucial to note that: (1) an object can be better characterized by leveraging comparative information between itself and other objects, and (2) an object's appearance can vary with camera position. As such, we present the Multi-view Approach to Grounding in Context (MAGiC) model, which selects an object referent based on language that distinguishes between two similar objects. By pragmatically reasoning over both objects and across multiple views of those objects, MAGiC improves over the state-of-the-art model on the SNARE object reference task with a relative error reduction of 12.9% (representing an absolute improvement of 2.7%). Ablation studies show that reasoning jointly over object referent candidates and multiple views of each object both contribute to improved accuracy.

Paper Type: long

Research Area: Multimodality and Language Grounding to Vision, Robotics and Beyond

Contribution Types: NLP engineering experiment

Languages Studied: English