The computer interrupted while Pamela was still speaking. I had accompanied her - my dear friend - to a recent doctor's appointment. She is in her 70s, lives alone while navigating multiple chronic health issues, and has been getting short of breath climbing the front stairs to her apartment. In the exam room, she spoke slowly and self-consciously, the way people often do when they are trying to describe their bodies and anxieties to strangers. Midway through her description of how she had been feeling, the doctor clicked his mouse and a block of text began to bloom across the computer monitor. The clinic had adopted an artificial-intelligence scribe, and it was transcribing and summarizing the conversation in real time.

Reasoning is a core capability in artificial intelligence systems, for which large language models (LLMs) have recently shown remarkable progress. However, most work focuses exclusively on deductive reasoning, which is problematic since other types of reasoning are also essential in solving real-world problems, and they are less explored. This work focuses on evaluating LLMs' inductive and abductive reasoning capabilities. We introduce a programmable and synthetic dataset, InAbHyD (pronounced in-a-bid), where each reasoning example consists of an incomplete world model and a set of observations. The task for the intelligent agent is to produce hypotheses to explain observations under the incomplete world model to solve each reasoning example. We propose a new metric to evaluate the quality of hypotheses based on Occam's Razor. We evaluate and analyze some state-of-the-art LLMs. Our analysis shows that LLMs can perform inductive and abductive reasoning in simple scenarios, but struggle with complex world models and producing high-quality hypotheses, even with popular reasoning-enhancing techniques such as in-context learning and RLVR.

Meta-learning stands for 'learning to learn' such that generalization to new tasks is achieved. Among these methods, Gradient-based meta-learning algorithms are a specific sub-class that excel at quick adaptation to new tasks with limited data. This demonstrates their ability to acquire transferable knowledge, a capability that is central to human learning. However, the existing meta-learning approaches only depend on the current task information during the adaptation, and do not share the meta-knowledge of how a similar task has been adapted before. To address this gap, we propose a 'Path-aware' model-agnostic meta-learning approach. Specifically, our approach not only learns a good initialization for adaptation, it also learns an optimal way to adapt these parameters to a set of task-specific parameters, with learnable update directions, learning rates and, most importantly, the way updates evolve over different time-steps. Compared to the existing meta-learning methods, our approach offers: (a) The ability to learn gradient-preconditioning at different time-steps of the inner-loop, thereby modeling the dynamic learning behavior shared across tasks, and (b) The capability of aggregating the learning context through the provision of direct gradient-skip connections from the old time-steps, thus avoiding overfitting and improving generalization. In essence, our approach not only learns a transferable initialization, but also models the optimal update directions, learning rates, and task-specific learning trends. Specifically, in terms of learning trends, our approach determines the way update directions shape up as the task-specific learning progresses and how the previous update history helps in the current update. Our approach is simple to implement and demonstrates faster convergence. We report significant performance improvements on a number of FSL datasets.

Pamela McCorduck is an artificial intelligence (AI) expert and author. She has written 10 books, the newest one comes out this month and it is titled: This Could Be Important: My Life and Times with the Artificial Intelligentsia. Pamela first became interested in AI when she was studying at the University of California, Berkeley. She was an English major, but she had a job typing in the business school, which is where she met some of the "fathers of AI". Two assistant professors in the business school who she became acquainted with through her job approached her to see if she could help them work on a book.

Pamela McCorduck is an artificial intelligence (AI) expert and an author. She has written 10 books, the newest one comes out this month and it is titled: This Could Be Important: My Life and Times with the Artificial Intelligentsia. Pamela first became interested in AI when she was studying at the University of California, Berkeley. She was an English major, but she had a job typing in the business school, which is where she met some of the "fathers of AI". Two assistant professors in the business school who she became acquainted with through her job approached her to see if she could help them work on a book.