Good vs Evil: Chosing the Right Moral Compass for you LLM

Mitigating bias in large language models is one of the most challenging and essential tasks in deep learning. But how do we define bias? What is "right" or "wrong," and how can we ensure our model’s moral compass aligns with the values of its intended community?

This no-nonsense talk dives into the complexities of bias, exploring the murky realities of data collection in an imperfect world and the expectations we place on our models. We’ll tackle sensitive topics head-on and explain, using intuitive concepts from elementary school geometry, what deep learning is truly learning—and how to control it.

Whether you're a researcher, practitioner, or simply curious, this session will equip you with a fresh perspective on the ethical landscape of AI and the tools to navigate it responsibly.