talks

Invited talks, workshop presentations, and lectures.

2026

  1. NLP Colloquium
    The Shape of Thought in Reasoning LLMs: A Reasoning Graph Perspective
    NLP Colloquium
    Feb 18, 2026
    Slides
  2. Matsuo Lab
    Analysis and Interpretability of LLMs
    Matsuo Lab LLM Course 2025 — Advanced Session 5
    Jan 7, 2026
    Slides

2025

  1. LLM-jp
    Large Language Models and Mechanistic Interpretability: Progress and Current State
    LLM-jp Mechanistic Understanding Workshop
    Aug 5, 2025
    Slides
  2. ICLR 2025
    In-Context Meta Learning Induces Multi-Phase Circuit Emergence
    ICLR 2025 Workshop on Building Trust in Language Models and Applications (Oral)
    Apr 24–28, 2025 · Singapore EXPO
    Slides