Based in Singapore. Interested in machine learning, language, and operations research.
- LeRobot dataset checks — A Python command-line tool for checking the integrity of LeRobot v2.1 datasets.
- Presupposed vs. asserted falsehoods — Experiments on how question framing affects LLM responses to false premises, with data and analysis code.
- Jane Street neural network puzzle — Code for reconstructing a shuffled residual network. Joint work with Keegan Ratner.
- Sichuan dialect in hip-hop lyrics — A coded lyric corpus and interview materials examining dialect in Chengdu hip-hop.