About
Hi, I’m Jinyeop Song, a final year PhD student in the MIT Physics Department. I am on the job market for Summer 2026. 
PhD candidate at MIT Physics researching the Science of AI (scaling laws, in-context learning, LLM agents) and AI for Science (foundation models for biology).
Always open to discussions and collaborations!
Research Interests
My research focuses on understanding the Science of AI, particularly on large language models (LLMs). Over the past few months, I’ve been exploring Neural Scaling Laws and In-Context Learning (ICL), uncovering how these models adapt and grow in capability.
Another key interest of mine is building AI for Science. I’m currently involved in projects developing Scalable Foundational Models for multiomics and proteins, helping advance scientific discovery through machine learning.
Featured Research
- Concept Encoding in In-Context Learning - ICML 2025 Spotlight
- Reconciling Kaplan and Chinchilla Scaling Laws - TMLR 2024
- EELMA: Estimating Empowerment of LLM Agents - AI Safety Research
News
- Nov 2025 - Featured in MIT EECS News for research on knowledge-graph augmented LLMs using multi-turn RL
- Sep 2025 - Two papers accepted to NeurIPS 2025 Workshops: Bayesian Optimization with LLMs and LLM Agent Empowerment
- Jun 2025 - Summer research intern at MIT-IBM Watson AI Lab
- May 2025 - Our paper accepted to ICML (Spotlight, Top 2.5%)! arXiv:2412.12276
- Jan 2025 - Started MATS 7.0 cohort with mentor Max Kleiman-Weiner (UW)
- Sep 2024 - MethylGPT paper under review at Nature Methods bioRxiv
- Aug 2024 - Scaling law paper accepted to TMLR arXiv:2406.12907
Contact
I’m always open to discussions and collaborations—feel free to reach out!
Location
Cambridge, MA, 02142