Full time - Researcher / Natural Language Processing (NLP) for Multimodal Machine Learning CVPR

Technology Field

Natural Language Processing

Machine Learning

Position Summary

Technologies like generative AI have the potential to transform the lifestyles of consumers and the workflows of professional creators. Sony R&D is developing large-scale generative AI technologies for content generation and restoration. This is called Sony generative AI, and is broken down into three categories: diffusion-based models, stochastic vector quantization techniques, and visual-and-language pre-training. Sony generative AI is expected to become an integral part of the music, film, and gaming industries in the years ahead. We at Sony R&D would like to make the most out of our unique opportunity to work directly with world-leading entertainment groups within those industries. Demonstrations of media generation and restoration are available at https://sony.github.io/creativeai.

Our team is building a wide range of natural language processing technologies, deep generative AI, and vision-language pre-training. You will work on advanced research on natural language processing and deep learning (e.g., vision-language pre-training, multimodal LLM).

Responsibilities

Fundamental research in natural language processing such as vision-language pre-training, multimodal learning, multimodal LLM, music/video understanding, agent, reasoning, controllable generative modeling, deep generative models for discrete data, image/audio captioning, text-to-image/audio, commonsense knowledge graphs, large-scale data development, etc.

You will be responsible for a wide range of activities, including paper submissions to top conferences (e.g. ACL, EMNLP, NeurIPS, ICLR, CVPR, etc.), collaborative research with universities and/or business groups, deployment of developed technologies within Sony and/or to third-party products together with product teams, etc.

You will also contribute to improving the efficiency of content creation in Sony's studios for music, movies, and game services by delivering AI-assisted tools developed by R&D.

Required qualifications

All of the following criteria are required.

■ Master's degree in natural language processing, artificial intelligence, machine learning, or closely related areas OR equivalent practical experience.

■ 3 years of experience with Python, C/C++, and Linux/Unix.

■ 2 years of experience in machine learning fields and NLP, using common frameworks such as PyTorch and TensorFlow.

■ Research ability, as demonstrated by a track record of conference papers, open-source software, or other scientific activities.

■ Ability to speak and write in English fluently and idiomatically.

Preferred qualifications

Qualifiction at the Ph.D level or higher in natural language processing, artificial intelligence, or machine learning is desirable.

Product, Service

Content creation support for movies/music/games, robots (Aibo), etc.

Development Environment

■OS: Windows and Linux

■Language: Python, C/C++, etc.

■PC, Server, Cloud Computing

Required Skills:

Optional Skills: