Multimedia Generative Script Learning for Task Planning
This work addresses the problem of assisting robots or humans in performing stereotypical activities by generating task plans, though it appears incremental as it builds on existing generative script learning with multimedia extensions.
The authors tackled the problem of generating subsequent steps for goal-oriented tasks by tracking historical states in both text and vision modalities, introducing a new benchmark with 5,652 tasks and 79,089 multimedia steps, and their approach significantly outperformed strong baselines in experiments.
Goal-oriented generative script learning aims to generate subsequent steps to reach a particular goal, which is an essential task to assist robots or humans in performing stereotypical activities. An important aspect of this process is the ability to capture historical states visually, which provides detailed information that is not covered by text and will guide subsequent steps. Therefore, we propose a new task, Multimedia Generative Script Learning, to generate subsequent steps by tracking historical states in both text and vision modalities, as well as presenting the first benchmark containing 5,652 tasks and 79,089 multimedia steps. This task is challenging in three aspects: the multimedia challenge of capturing the visual states in images, the induction challenge of performing unseen tasks, and the diversity challenge of covering different information in individual steps. We propose to encode visual state changes through a selective multimedia encoder to address the multimedia challenge, transfer knowledge from previously observed tasks using a retrieval-augmented decoder to overcome the induction challenge, and further present distinct information at each step by optimizing a diversity-oriented contrastive learning objective. We define metrics to evaluate both generation and inductive quality. Experiment results demonstrate that our approach significantly outperforms strong baselines.