Tian Huey Teh
Source-listed: Google DeepMind
Record snapshot: Sep 3, 2026. Coverage and affiliations may be incomplete or historical.
Selected work
5- Debate Training Reduces Reward Hacking in RLAIFSaved credit: Contributor · 2026Explore people connected to this work →
- Decoding Safety Feedback from Diverse Raters: A Data-driven Lens on Responsiveness to SeveritySaved credit: Contributor · 2026Explore people connected to this work →
- Whose View of Safety? A Deep DIVE Dataset for Pluralistic Alignment of Text-to-Image ModelsSaved credit: Contributor · 2025Explore people connected to this work →
- Human-AI Complementarity: A Goal for Amplified OversightSaved credit: Contributor · 2025Explore people connected to this work →
- Gemini: A Family of Highly Capable Multimodal ModelsSaved credit: Contributor · 2023Explore people connected to this work →