Bias in Language Models
Tim and Minqi discuss how pruning language models introduces bias by favoring human preferences, impacting the diversity of generated answers. The choice of humans providing preference data plays a crucial role in shaping the model's output.In this clip
From this podcast
Machine Learning Street Talk (MLST)
#114 - Secrets of Deep Reinforcement Learning (Minqi Jiang)
Related Questions
How does this language model work?
What role do biases play in learning as discussed in the episode Language Understanding and LLMs with Christopher Manning - 686 and the clip Learning with Bias?
How are Large Language Models (LLMs) fine-tuned post-training in the episodes Is ChatGPT Getting Worse? with James Zou - 645 and Medical Language Models?