This socialbot's vision is to create a bot that conducts engaging and highly personalized conversations with users.
Our vision is to create a socialbot that conducts engaging and highly personalized conversations with users. Our philosophy is that every part of the conversation should give us some new information about the user's personality or preferences. In our vision, Alquist will find relevant context knowledge (e.g., an article about football) based on a current conversation. Next, Alquist will pick up a relevant piece of information such that the conversation will be meaningful and coherent (e.g., the conversational agent should not contradict itself).
Ondřej Kobza - Team leader
Kobza is a PhD student of artificial intelligence at Czech Technical University in Prague. His primary focus is conversational AI, namely finding relevant information in text (information retrieval and question answering) and natural language generation w.r.t. knowledge. His interest in NLP began when he was on ERASMUS+ at KU Leuven. He joined E- Club (NLP research group at CTU) afterwards, where he wrote my master's thesis focused on generating natural sentences from RDF triples. Later, he became part of the winning team Alquist during the Alexa Prize Socialbot Grand Challenge 4.
Čuhel is a master's student in computer vision at the Czech Technical University (CTU). He is interested in NLP and CV. Currently, he is discovering the potential of generative language models in paraphrasing and question generation tasks. His goal is to use these models for enlarging training data sets. Previously, Čuhel worked in the domain of Information retrieval, where he trained several models for re-ranking relevant documents with respect to a query. He earned his bachelor's from FEE CTU in computer science with a specialization in AI. His bachelor's thesis focused on multimodal emotion recognition from text, audio, and a combination of both.
Gargiani is pursuing his master's in data science with an interest in NLP and its innovative applications. Gargiani's bachelor's thesis focused on intent classification and out-of-domain (outlier) detection, where his proposed method outperformed the current SOTA. Now, Gargiani works with language models and explores their capabilities as knowledge bases and use them to observe current world trends.
Herel is PhD student in the computer science doctoral study program at the Department of Cybernetics, Faculty of electrical Engineering (FEE), Czech Technical University (CTU). He is researching novel methods in Natural Language Processing under the supervision of Tomas Mikolov. Previously, he earned his bachelor's and master’s degree from FEE CTU in computer science. Before starting doctoral studies, Herel was a junior researcher at Czech Institute of Informatics, Robotics and Cybernetics (CIIRC), CTU, where he focused on emergence of novelty in evolutionary algorithms.
Marek is a PhD student of artificial intelligence at Czech Technical University in Prague. His primary focus is conversational AI, neural response generators and dialogue management. He was developing Alexa Prize winning socialbot Alquist during the Alexa Prize Socialbot Grand Challenges 1, 2, 3 and 4. He was interning at Amazon Lab126 as an Applied Scientist in summer 2019 and 2020. He also volunteers as a scoutmaster.
Jan Šedivý - Faculty advisor
Šedivý has three decades of experience in the IT industry. He has led numerous global research and development projects and is the holder of 19 US patents. He began as a researcher and research manager at the IBM Thomas J. Watson Research Center (1992-2008), later moving to Google as a Technical Lead Manager (2008-2010). He subsequently returned to CTU-CIIRC, where he leads the NLP group - which has won the Amazon Alexa in 2021.