Model Challenges & Bias
Tatsu discusses the difficulty models face in long-range planning tasks, shedding light on limitations in current language models. Patrick delves into the nuances of bias in language models, exploring how opinions and values are reflected in their outputs, raising thought-provoking questions about model training and societal influences.In this clip
From this podcast
Unsupervised Learning
Ep 11: Stanford Professor Tatsu Hashimoto on AI Biases and Improving LLM Performance
Related Questions