3 papers
cs.CL2025
Uncovering Spontaneous Physics Representations in In-Context Learning
Yeongwoo Song, Jaeyong Bae, Dong-Kyum Kim +1
In-context learning (ICL) lets large language models (LLMs) solve new tasks from prompts alone, across an ever-widening range of domains, yet the mechanisms underlying this ability…
cs.LG2025
Stochastic Resetting Mitigates Latent Gradient Bias of SGD from Label Noise
Youngkyoung Bae, Yeongwoo Song, Hawoong Jeong
Giving up and starting over may seem wasteful in many situations such as searching for a target or training deep neural networks (DNNs). Our study, though, demonstrates that resett…
cs.LG2024
Towards Cross Domain Generalization of Hamiltonian Representation via Meta Learning
Yeongwoo Song, Hawoong Jeong
Recent advances in deep learning for physics have focused on discovering shared representations of target systems by incorporating physics priors or inductive biases into neural ne…