Jianyang GaoAboutAI ResearcherSpeaking appearancesPaper Reading Group | Linear Representation Hypothesis in the context of LLM Safety 2026-05-14 — Zürich