CURIE: An Iterative Querying Approach for Reasoning About Situations

TL;DR

CURIE introduces an iterative method using finetuned language models to build structured situational graphs that enhance reasoning about new situations across various domains.

Contribution

It presents a novel approach to explicitly construct situational graphs via natural language queries, improving reasoning accuracy in complex tasks.

Findings

Generated st-graphs are relevant and meaningful to humans.

Augmenting reasoning tasks with st-graphs improves accuracy by 3 points.

Particularly effective for tasks requiring background knowledge and multi-hop reasoning.

Abstract

Recently, models have been shown to predict the effects of unexpected situations, e.g., would cloudy skies help or hinder plant growth? Given a context, the goal of such situational reasoning is to elicit the consequences of a new situation (st) that arises in that context. We propose a method to iteratively build a graph of relevant consequences explicitly in a structured situational graph (st-graph) using natural language queries over a finetuned language model (M). Across multiple domains, CURIE generates st-graphs that humans find relevant and meaningful in eliciting the consequences of a new situation. We show that st-graphs generated by CURIE improve a situational reasoning end task (WIQA-QA) by 3 points on accuracy by simply augmenting their input with our generated situational graphs, especially for a hard subset that requires background knowledge and multi-hop reasoning.