University of Cambridge > > NLIP Seminar Series > Adaptation and Control in Enterprise Language Technology

Adaptation and Control in Enterprise Language Technology

Add to your list(s) Download to your calendar using vCal

  • UserRyan McDonald (ASAPP)
  • ClockFriday 07 May 2021, 12:00-13:30
  • HouseVirtual (Zoom).

If you have a question about this talk, please contact Georgi Karadzhov.

This talk will focus on NLP models in an enterprise setting. Unlike consumer settings, enterprise solutions require NLP to rapidly adapt to new domains as well as provide some level of control for the user, either dynamically or statically. This is to facilitate scale with respect to the number of distinct clients of the service. Scale in enterprise solutions can only be achieved through the development of ML and NLP solutions that minimize the need for ML/NLP practitioners to be involved in every new use case. To this end, motivated by assisted document writing, I will discuss two recent papers that focus on adaptation through synthetic data creation and control in generative models, the former applied to retrieval and the latter to summarization. I will show that simple techniques can help automate the construction of models for both of these challenges making deployment seamless.

This is based on work with colleagues from Google and ASAPP .

Join Zoom Meeting

Meeting ID: 971 0677 0495 Passcode: 038254

This talk is part of the NLIP Seminar Series series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2024, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity