Multi-Modal and Multi-Agent Systems Meet Rationality: A Survey
Abstract: Rationality is the quality of being guided by reason, characterized by decision-making aligned with evidence and logical rules. This quality is essential for effective problem-solving, as it ensures that solutions are well-founded and consistently derived. Despite the advancements of large language models (LLMs) in generating human-like texts with remarkable accuracy, they present limited knowledge space, inconsistency across contexts, and difficulty understanding complex scenarios. Therefore, recent research focuses on building multi-modal and multi-agent systems to achieve considerable progress with enhanced consistency and reliability, instead of relying on a single LLM as the sole planning or decision-making agent. To that end, this paper aims to understand whether multi-modal and multi-agent systems are advancing toward rationality by surveying the state-of-the-art works, identifying advancements over single-agent and single-modal systems in terms of rationality, and discussing open problems and future directions.
Paper Type: Long
Research Area: Multimodality and Language Grounding to Vision, Robotics and Beyond
Research Area Keywords: rationality, multi-agent systems, multimodality, evaluation methodologies, factuality, multimodal applications
Contribution Types: Surveys
Languages Studied: English
Submission Number: 4201
Loading