Koshiro Aoki

Hello, I'm Koshiro Aoki!

I am a master’s student in Computer Science at Waseda University. I like thinking about thinking. This has several meanings: metacognition (thinking about one’s own thinking), interpretability (thinking about AI’s thinking), and NeuroAI (think about the brain’s thinking through AI).

CV

Selected Publications

Bottom-Up Interpretability of Pretraining Dynamics via Loss Curve Decomposition

Bottom-Up Interpretability of Pretraining Dynamics via Loss Curve Decomposition

Koshiro Aoki, Masaru Isonuma, Max Müller-Eberstein, Yusuke Oda, Hirokazu Kiyomaru, Takashi Kodama, Chaoran Liu, Yohei Oseki, Yusuke Miyao, Daisuke Kawahara

Third Conference on Language Modeling (COLM 2026)

Crosscoders Identify Shared or Specific Features between the Human Brain and Language Models

Crosscoders Identify Shared or Specific Features between the Human Brain and Language Models

Koshiro Aoki, Itsuki Hamada, Naho Orita, Daisuke Kawahara, Hiromu Sakai

Mechanistic Interpretability Workshop @ ICML 2026

In-Context Neurofeedback: Can Large Language Models Control Their Internal Representations through Privileged Access?

In-Context Neurofeedback: Can Large Language Models Control Their Internal Representations through Privileged Access?

Koshiro Aoki, Ryota Takatsuki, Gouki Minegishi, Yusuke Haruki, Daisuke Kawahara

Trustworthy AI for Good (AI4GOOD) Workshop @ ICML 2026 Spotlight

Others