A Picture Is Worth a Graph: A Blueprint Debate Paradigm for Multimodal Reasoning
This paper presents a pilot study aimed at introducing multi-agent debate into multimodal reasoning. The study addresses two key challenges: the trivialization of opinions resulting from excessive summarization and the diversion of focus caused by distractor concepts introduced from images. These challenges stem from the inductive (bottom-up) nature of existing debating schemes. To address the issue, we propose a deductive (top-down) debating approach called Blueprint Debate on Graphs (BDoG). In BDoG, debates are confined to a blueprint graph to prevent opinion trivialization through world-level summarization. Moreover, by storing evidence in branches within the graph, BDoG mitigates distractions caused by frequent but irrelevant concepts. Extensive experiments validate that BDoG is able to achieve state-of-the-art results in ScienceQA and MMBench with significant improvements over previous methods. The source code can be accessed at https://github.com/thecharm/BDoG.
Authors
- Xiao-Yong Wei (ORCID: https://orcid.org/0000-0002-5706-5177)
- Changmeng Zheng (ORCID: https://orcid.org/0000-0002-2945-8248)
- Tat‐Seng Chua (ORCID: https://orcid.org/0000-0001-6097-7807)
- Qing Li (ORCID: https://orcid.org/0000-0003-3370-471X)
- Dayong Liang
- Wengyu Zhang (ORCID: https://orcid.org/0009-0001-2347-4183)
Institutions
- Hong Kong Polytechnic University (HK)
- National University of Singapore (SG)
- South China University of Technology (CN)
Publication Details
- Published
- 2024-10-26
- DOI
- https://doi.org/10.1145/3664647.3681102
- Citations
- 2
- Primary Topic
- Multimodal Machine Learning Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.50