Multimodal Procedural Planning via Dual Text-Image Prompting

Yujie Lu, Pan Lu, Zhiyu Chen, Wanrong Zhu, Xin Eric Wang, William Yang Wang

Keywords: Multimodal Procedural Planning, Large Language Model, Text-to-Image Synthesis

TL;DR: Our dual Text-Image Prompting (TIP) model generates coherent and authentic multimodal procedural plans towards a high-level goal, providing useful guidelines in task completion.

Abstract: Embodied agents have achieved prominent performance in following human instructions to complete tasks. However, the potential of providing instructions informed by texts and images to assist humans in completing tasks remains unexplored. To uncover this capability, we present the multimodal procedural planning (MPP) task, in which models are given a high-level goal and generate plans of paired text-image steps, providing more complementary and informative guidance than unimodal plans. The key challenges of MPP are to ensure the informativeness, temporal coherence, and accuracy of plans across modalities. To tackle this, we propose Text-Image Prompting (TIP), a dual-modality prompting method that jointly leverages zero-shot reasoning ability in large language models (LLMs) and compelling text-to-image generation ability from diffusion-based models. Our TIP improves the interaction in the dual modalities using text-to-image bridge and image-to-text bridge, allowing LLMs to guide the textual-grounded image plan generation and leveraging the descriptions of image plans to ground the textual plan reversely. To address the lack of relevant datasets, we collect WIKIPLAN and RECIPEPLAN as a testbed for MPP. Our results show compelling human preferences and automatic scores against unimodal and multimodal baselines on \wiki~ and \recipe~ in terms of informativeness, temporal coherence, and plan accuracy.

Primary Area: generative models

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 7941