Publication
Chain-of-Thought Oversight Should Not Treat Faithfulness as Monitorability accepted at NeurIPS 2026.
N02 / Academic portfolio
News
Appointments, paper acceptances, and current directions across AI governance, explainability, and interdisciplinary applications.
News / Current
Latest updates.
Confirmed announcements from the current site record.
-
-
Publication
NoRA: Evaluating Grounded Reasonableness in Visual First-person Normative Action Reasoning accepted at NeurIPS 2026.
-
Workshop
Education in the Age of AI: Human-Centred Learning, Teaching, and Assessment with Generative AI (EdAI 2026) will be held at AJCAI 2026.
-
Publication
Student-Centered AI Requires Auditable Learning Evidence accepted at AACL-IJCNLP 2026.
-
Publication
Ownership Tracing in AI Tutoring: Correcting Apparent Mastery with Source-of-Practice Evidence accepted at ICDM 2026.
-
Appointment
Joined The University of Sydney as Lecturer in Computer Science.
-
Publication
Practical Attribution Guidance for Rashomon Sets accepted at AISTATS.
-
Publication
FaithAct accepted at ACL.
-
Research
Recent work on Rashomon-aware multimodal learning for battery materials is forthcoming in Advanced Intelligent Discovery.
-
Projects
Ongoing projects explore explainable LLMs, AI tutoring, and trustworthy AI for science.
Academic portfolio