I am fascinated by the intersection of artificial intelligence and human intelligence. I want to know how intelligence works through the lenses of deep learning:
Geometries of learned representation spaces in deep learning, esp. LLMs and VLMs
Generative models as the approximation of reality
World models in language models and model-based reinforcement learning
- Interpretability
- Language Models
- Model-Based Reinforcement Learning
I wish this list talks about my research interests.
🔎 The Platonic Representation Hypothesis (Huh et al., 2024)
🌏 Language Models Represent Space and Time (Gurnee and Tegmark, 2024)
💎 Mastering Diverse Domains through World Models (Hafner et al., 2023)