Xiangchen Song (宋祥辰)
I am a PhD student in Machine Learning Department at Carnegie Mellon University, advised by Prof. Kun Zhang (CMU-CLeaR Group). Previously, I studied Computer Science at UIUC with Prof. Jiawei Han.
research
Large language models are sequence models, and my work aims to make their internal representations provably identifiable so we can interpret and steer model behavior with principled guarantees. I build on causal representation learning to recover latent structure in temporal data (time series, video, and text), and bring this lens to mechanistic interpretability of LLMs: designing identifiable sparse autoencoders with feature consistency for LLM activations, analyzing internal reasoning mechanisms, and enabling targeted control for more reliable and efficient model behavior.
contact
Email: xiangchs [at] cs [dot] cmu [dot] edu
news
| Apr 08, 2026 | I received a Modal for Academics compute grant to support my research on LLM test-time training. Many thanks to Modal for their generous support! |
|---|---|
| Apr 06, 2026 | One paper on mechanistic interpretability and one paper on diffusion large language models have been accepted to The 64th Annual Meeting of the Association for Computational Linguistics (ACL’2026)!! |
| Sep 23, 2025 | Two papers about efficient LLM reasoning have been accepted to NeurIPS 2025 Workshop on Efficient Reasoning (ER@NeurIPS’2025)!! |
| Sep 22, 2025 | Two papers on mechanistic interpretability have been accepted to Mechanistic Interpretability Workshop at NeurIPS 2025 (MechInterp@NeurIPS’2025)!! |
| Sep 18, 2025 | One paper “LLM Interpretability with Identifiable Temporal-Instantaneous Representation” has been accepted to The Thirty-ninth Conference on Neural Information Processing Systems (NeurIPS’2025)!! |