Defining Reasoning
Yasaman and Sameer discuss the challenges in defining reasoning in machine learning. Yasaman emphasizes the importance of robustness and following mental models, while Sameer highlights the need for models to understand inputs contextually rather than rely solely on pre-training data. They explore examples like operation inference tests to illustrate what true reasoning might entail.In this clip
From this podcast
Machine Learning Street Talk (MLST)
#73 - YASAMAN RAZEGHI & Prof. SAMEER SINGH - NLP benchmarks
Related Questions