Large Model Application Algorithm Research Scientist-International Content Security Algorithm Research-Soaring Star Talent Program

Tasks

• Apply reinforcement learning to natural language tasks
• Design reward models for reinforcement learning
• Develop large language models
• Evaluate model reasoning performance
• Improve reasoning efficiency
• Monitor and mitigate content risks
• Train and stabilize reinforcement learning without supervised fine tuning

Perks/Benefits

• N/A

Skills/Tech-stack

Chain-of-Thought | Data Compliance | Fine Tuning | Knowledge Distillation | Language Models | Language Processing | Large Language Models | Machine Learning | Model Evaluation | Monte Carlo | Monte-Carlo Tree Search | Natural Language | Natural Language Processing | Process-based Reward Model | Reinforcement Learning | Reward Model | Supervised Fine Tuning | Tree search