A. Lapedriza

Latest

• Human-Centric Dialog Training via Offline Reinforcement Learning
• Approximating Interactive Human Evaluation with Self-Play for Open-Domain Dialog Systems