Draw Me a Flower: Processing and Grounding Abstraction in Natural Language

TL;DR

This paper introduces a new method and dataset for studying how natural language processing models understand and ground abstractions in instructions, revealing current models' limitations especially with higher abstraction levels.

Contribution

The work presents a novel abstraction elicitation method, a new dataset of visually-grounded instructions, and an evaluation framework highlighting the challenges models face with abstraction.

Findings

Models perform worse than humans on abstraction tasks.

Performance declines as abstraction level increases.

Contemporary models show significant room for improvement.

Abstract

Abstraction is a core tenet of human cognition and communication. When composing natural language instructions, humans naturally evoke abstraction to convey complex procedures in an efficient and concise way. Yet, interpreting and grounding abstraction expressed in NL has not yet been systematically studied in NLP, with no accepted benchmarks specifically eliciting abstraction in NL. In this work, we set the foundation for a systematic study of processing and grounding abstraction in NLP. First, we deliver a novel abstraction elicitation method and present Hexagons, a 2D instruction-following game. Using Hexagons we collected over 4k naturally-occurring visually-grounded instructions rich with diverse types of abstractions. From these data, we derive an instruction-to-execution task and assess different types of neural models. Our results show that contemporary models and modeling practices are substantially inferior to human performance, and that models' performance is inversely correlated with the level of abstraction, showing less satisfying performance on higher levels of abstraction. These findings are consistent across models and setups, confirming that abstraction is a challenging phenomenon deserving further attention and study in NLP/AI research.