Saved in:
Bibliographic Details
Main Authors: Tabassum, Afrina, Guo, Bin, Ma, Xiyao, Eldardiry, Hoda, Lourentzou, Ismini
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2509.21662
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908559284895744
author Tabassum, Afrina
Guo, Bin
Ma, Xiyao
Eldardiry, Hoda
Lourentzou, Ismini
author_facet Tabassum, Afrina
Guo, Bin
Ma, Xiyao
Eldardiry, Hoda
Lourentzou, Ismini
contents Multimodal Procedural Planning (MPP) aims to generate step-by-step instructions that combine text and images, with the central challenge of preserving object-state consistency across modalities while producing informative plans. Existing approaches often leverage large language models (LLMs) to refine textual steps; however, visual object-state alignment and systematic evaluation are largely underexplored. We present MMPlanner, a zero-shot MPP framework that introduces Object State Reasoning Chain-of-Thought (OSR-CoT) prompting to explicitly model object-state transitions and generate accurate multimodal plans. To assess plan quality, we design LLM-as-a-judge protocols for planning accuracy and cross-modal alignment, and further propose a visual step-reordering task to measure temporal coherence. Experiments on RECIPEPLAN and WIKIPLAN show that MMPlanner achieves state-of-the-art performance, improving textual planning by +6.8%, cross-modal alignment by +11.9%, and visual step ordering by +26.7%
format Preprint
id arxiv_https___arxiv_org_abs_2509_21662
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MMPlanner: Zero-Shot Multimodal Procedural Planning with Chain-of-Thought Object State Reasoning
Tabassum, Afrina
Guo, Bin
Ma, Xiyao
Eldardiry, Hoda
Lourentzou, Ismini
Machine Learning
Multimodal Procedural Planning (MPP) aims to generate step-by-step instructions that combine text and images, with the central challenge of preserving object-state consistency across modalities while producing informative plans. Existing approaches often leverage large language models (LLMs) to refine textual steps; however, visual object-state alignment and systematic evaluation are largely underexplored. We present MMPlanner, a zero-shot MPP framework that introduces Object State Reasoning Chain-of-Thought (OSR-CoT) prompting to explicitly model object-state transitions and generate accurate multimodal plans. To assess plan quality, we design LLM-as-a-judge protocols for planning accuracy and cross-modal alignment, and further propose a visual step-reordering task to measure temporal coherence. Experiments on RECIPEPLAN and WIKIPLAN show that MMPlanner achieves state-of-the-art performance, improving textual planning by +6.8%, cross-modal alignment by +11.9%, and visual step ordering by +26.7%
title MMPlanner: Zero-Shot Multimodal Procedural Planning with Chain-of-Thought Object State Reasoning
topic Machine Learning
url https://arxiv.org/abs/2509.21662