Incorporating Task-specific Concept Knowledge into Script Learning

TL;DR

This paper introduces Tetris, a goal-oriented script completion task that incorporates user context, proposing concept prompting and contrastive learning techniques to improve performance, with promising results on a WikiHow-based dataset.

Contribution

It presents a new task of script completion considering user context and introduces novel techniques like concept prompting and contrastive learning.

Findings

Both proposed methods improve performance on the WikiHow dataset.

The dataset, models, and code will be publicly available.

Addressed step repetition and hallucination issues in script learning.

Abstract

In this paper, we present Tetris, a new task of Goal-Oriented Script Completion. Unlike previous work, it considers a more realistic and general setting, where the input includes not only the goal but also additional user context, including preferences and history. To address this problem, we propose a novel approach, which uses two techniques to improve performance: (1) concept prompting, and (2) script-oriented contrastive learning that addresses step repetition and hallucination problems. On our WikiHow-based dataset, we find that both methods improve performance. The dataset, repository, and models will be publicly available to facilitate further research on this new task.