Gregory Kell
greg-huawei
AI & ML interests
None yet
Recent Activity
upvoted a paper about 1 month ago
Memento 3: Model-Based Recursive Self-Improvement through Reflective Rulebooks upvoted a paper about 1 month ago
Beam Search as Test-Time Self-Distillation via Counterfactual Contexts upvoted a paper about 1 month ago
The Weakest Link: Distilling LLM Reasoning with Worst-Case Constrained Reinforcement LearningOrganizations
None yet