Postdoctoral Associate, Convo AI, Personal Robots
Job Description
POSTDOCTORAL ASSOCIATE, CONVO AI, PERSONAL ROBOTS, Program in Media Arts and Sciences, to join the Personal Robots research group. Will bring experience building large language models (LLMs) and specialized conversational systems to support next-generation human experience over long-term encounters with personified social robots; conduct independent and collaborative research in the conversational AI field; publish conference and journal research papers; make public presentations; collaborate on designing and developing innovative conversational AI systems such as embodied conversational agents; collect data sets for training new models; design evaluation studies and collect and analyze data for research projects; participate in grant-writing and fundraising; mentor/advise graduate students and junior researchers; and stay up-to-date with the latest research and trends in the field.
A full job description is available here.
REQUIRED: completed Ph.D. thesis in neural conversational systems or closely related area; publications in conversational AI venues (e.g., EMNLP, ACL, NAACL, IJCAI, ICLR, NeurIPS, AAAI, etc.); understanding of human-in-the-loop NLP and human-centered tasks; excellent coding skills in modeling and conversational interface design for real-time interaction (e.g., PyTorch/TensorFlow and Python proficiency); and experience/interest in deploying conversational systems in real-world applications. PREFERRED: interest/experience in the following--multimodal connections to generative language models, personalized LLM, persona-based LLM, and/or LLM with/for RL planning; familiarity with the MIT RAISE Initiative’s work, including the MIT App Inventor and Personal Robotics Group, ethical or interactive limitations in current LLM systems, social emotional reasoning using language, language grounding in social interactions, multi-person dialog, bilingual/multilingual problems (cultural, linguistics and interaction, and neuro-symbolic or knowledge graph-based systems; and interest in developing graduate-level seminars/courses to teach the design, computational approaches, development, and ethical usage with neural-based conversational agents. Job #22542
This is a one-year appointment with the possibility of extension due to funding and course of the research.
3/28/23
*Please mention you saw this ad on WomenInTheSciences.*