Language Technologies Institute at Carnegie Mellon University

The "Let's Go" dataset, consisting of 171,128 real-world interactions between real users collected in a project led by LTI faculty members Maxine Eskenazi and Alan Black, has been released in its entirety: https://github.com/DialRC/LetsGoDataset Use it as a resource to train/test your dialog models!

https://www.geekwire.com//top-schools-ai-new-study-ranks-/

Members of @LTICMU have a large number of papers coming up at #AAAI2018 next week, be sure to check them out! "A Continuous Relaxation of Beam Search for End-to-end Training of Neural Sequence Models" (Goyal et al.) https://t.co/SwJAwDFW8Q 1/

Today's @LTICMU colloquium is Kate Saenko on "Why did you say that? Explaining neural models of vision and language"! https://t.co/j3WmFkbEYK 2:30PM at Doherty Hall 2315.

Retweeted CMU Computer Science (@SCSatCMU): Tartan team from @LTICMU @SCSatCMU is one of 8 sponsored teams worldwide chosen to create a socialbot to compete in the $3.5M Amazon #AlexaPrize https://t.co/41Xo5ZCyu1