Commonsense Reasoning
Commonsense reasoning, the ability of AI systems to understand and apply everyday knowledge, is a crucial area of research aiming to bridge the gap between human and artificial intelligence. Current research focuses on integrating large language models (LLMs) with other modalities like vision and tactile data, often using techniques like instruction tuning, multimodal learning, and knowledge graph integration to improve performance on various benchmarks. This work is significant because enhanced commonsense reasoning is essential for building more robust, reliable, and explainable AI systems across diverse applications, including robotics, deepfake detection, and conversational AI.
Papers
UNcommonsense Reasoning: Abductive Reasoning about Uncommon Situations
Wenting Zhao, Justin T Chiu, Jena D. Hwang, Faeze Brahman, Jack Hessel, Sanjiban Choudhury, Yejin Choi, Xiang Lorraine Li, Alane Suhr
SAIE Framework: Support Alone Isn't Enough -- Advancing LLM Training with Adversarial Remarks
Mengsay Loem, Masahiro Kaneko, Naoaki Okazaki