To develop and implement novel deep learning/reinforcement learning algorithms for natural language processing to enable fast, accurate and safe interactions on Samsung devices.
Requirements
- Strong background in machine learning (supervised learning, transfer learning, one-shot or few shot learning, unsupervised learning, semi-supervised learning, weakly supervised learning, meta-learning, outlier detection, etc.)
- Expertise in LLM including model architecture, training/finetuning techniques, retrieval augmented generation (RAG), reasoning and action planning, etc.
- Experience in conversational AI technologies: natural language processing (e.g., language models, semantic parsing, natural language generation etc.), dialogue (e.g., state tracking, policy learning), and representation learning (embedding, conceptualization, etc.)
- Proficiency in a neural network library (e.g., PyTorch, TensorFlow)
- Experience in on-device AI technologies such as lightweight model architecture design
- Experience in multimodal AI technologies for various multimodal applications
Responsibilities
- Develop and implement novel deep learning/reinforcement learning algorithms for natural language processing
- Contribute to the research activities of the team
Other
- Teamwork and communication skills
- Current Ph.D. student in CS, EE, or related field
- Ability to work in an office setting, sit and stand at a desk, communicate in person and by telephone, and frequently operate standard office equipment, such as telephones and computers
