Path Breaking Emergence
Path-breaking emergence in artificial intelligence focuses on understanding how complex, unexpected capabilities arise from the interactions of simpler components within large-scale models, particularly deep neural networks and large language models (LLMs). Current research investigates this phenomenon through various lenses, including analyzing training dynamics, exploring the role of model architecture (e.g., transformers, recurrent networks), and developing quantitative metrics to measure emergence. These studies aim to improve our understanding of model behavior, enhance model design and training, and ultimately contribute to safer and more reliable AI systems.
Papers
The Emergence of Reproducibility and Generalizability in Diffusion Models
Huijie Zhang, Jinfan Zhou, Yifu Lu, Minzhe Guo, Peng Wang, Liyue Shen, Qing Qu
The Troubling Emergence of Hallucination in Large Language Models -- An Extensive Definition, Quantification, and Prescriptive Remediations
Vipula Rawte, Swagata Chakraborty, Agnibh Pathak, Anubhav Sarkar, S. M Towhidul Islam Tonmoy, Aman Chadha, Amit P. Sheth, Amitava Das