Be the first to try a brand new demo of DeepPavlov Library!

DREAM Technical Report 2019
DREAM Socialbot filled the gap between production-ready agents and state-of-the-art NLP research. DREAM Socialbot achieved final average rating of 3.39.
After the competition, we've built an AI assistant based on the our DREAM socialbot.

We call it DeepPavlov Dream. You can think of it as a first distribution of an open source multiskill AI assistant like Linux but for Conversational AI.

In September 2020, we've published a running demo of this distribution, as well as a number of simple multiskill AI Assistants derived from this distribution.

The running demo has a number of our default goal-oriented and chit-chat skills as well as the supporting annotators from Library. It is orchestrated by DeepPavlov Agent. You can talk to it on our Demo Website and in Telegram Messenger.
You can play with the smaller multiskill AI assistants built using DeepPavlov Dream in our demos repo.
  • Yury Kuratov
    Team Leader
    Yury is a Ph.D. student and researcher since 2017. Previously Yury worked as a software developer at ABBYY on building Russian Corpora for three years. He has a master degree in Computational Linguistics. Yury's research area is neural networks for natural language processing. His main projects are coreference resolution, question answering and language modeling. Yury and Idris won Conversational Intelligence Challenge (ConvAI) in 2017 and have publications at NIPS and COLING conferences. Yury has strong skills in research, development and team management.
  • Diliara Baymurzina
    Diliara is a Ph.D. student and researcher. She has two master degrees in Computational Mathematics. Diliara's current research is dedicated to neural approaches to text classification and evolutionary algorithms for neural architecture search. One of her main responsibilities in the DeepPavlov project is development of text classification components.
  • Idris Yusupov
    Idris is a Ph.D. student and researcher. Idris has developed various systems from web-apps on Ruby on Rails to dialog agents that uses wide range of NLP and machine learning methods. Currently, he is leading the development of NLP-core in an chatbot platform. Idris is a Ph.D. Student focusing his area of research is dialog agents. Idris and Yury won Conversational Intelligence Challenge 1 (ConvAI) and have publications in NIPS and COLING conferences. He is a team player, that gets things done.
  • Anh Le
    Anh is a Ph.D. student and researcher. In 2012 he received his Masters degree in Computer Science from the University of Engineering and Technology, Vietnam National University, Hanoi, Viet Nam. He is a lecturer at the Faculty of Information Technology - Vietnam Maritime University. The Anh's research is related to deep neural network models for natural language processing tasks and focuses on structural retrieval from text (e.g. named entity recognition, coreference resolution).
  • Denis Kuznetsov
    Denis is a Ph.D. student and researcher since 2019. Since 2016, Denis began to actively engage in the development of models on neural networks. Denis's research is focused on neural machine translation, neural networks attention mechanisms with constrained condition, the bulk of research is generative models and their improvement.
  • Elena Ermakova
    Elena is a MS student and researcher since 2018. Currently she is receiving master's degree in Data Science. Elena's research is focused on retrieval methods applied to different tasks e.g. chit-chat conversation and Q&A with explanation.
  • Dmitry Karpov
    Dmitry is a Ph.D. student and researcher since 2019. Dmitry won the Moscow chess championship among private school students. During study at MIPT he has received 2 extra scholarships (Abramov scholarship, state elevated academic scholarship), and earned his BA degree with honors. In 2017, Dmitry switched his research area to Computer Science. Since then he has managed to win the DeepHack.Chat competition.
  • Daniil Cherniavskii
    Daniil is a bachelor student and researcher at his final year. His research is focused on neural networks attention and memory mechanism in natural language processing.
  • Mikhail Burtsev
    Faculty Advisor
    Mikhail Burtsev is a head of DeepPavlov project. In 2005, he received the Ph.D. degree from Keldysh Institute of Applied Mathematics of Russian Academy of Sciences. He is one of the organizers of NeurIPS Conversational Intelligence Challenges (ConvAI 1&2) in 2017 and 2018, as well as a series of workshops on Search-oriented Conversational AI. At the moment he leads the development of open-source conversational AI framework DeepPavlov.
We are the team from the Neural Networks and Deep Learning Lab at Moscow Institute of Physics and Technology. The main research area of our lab is NLP with a focus on Conversational Intelligence and Dialog Agents. We develop a conversational artificial intelligence framework DeepPavlov that contains all the components required for building chatbots.