Seonglae Cho- Sparse Mixture‑of‑Experts(MoE)
- Thinking RL
arxiv.org
https://arxiv.org/pdf/2507.06261
Deep Think
Try Deep Think in the Gemini app
Deep Think utilizes extended, parallel thinking and novel reinforcement learning techniques for significantly improved problem-solving.
https://blog.google/products/gemini/gemini-2-5-deep-think/
