Search results

Applied Scientist

Privacy is an important concern when building statistical models on data containing personal information. Differential privacy offers a strong definition of privacy and can be used to solve several privacy concerns (Dwork et al., 2014). Multiple solutions have been proposed for the differentially-private transformation of datasets containing sensitive information. However, such transformation algorithms

Principal, Applied Scientist

Machine learning (ML) is increasingly used to automate decision making in various domains. Almost all common ML models are susceptible to data errors in the serving data (for which the model makes predictions). Such errors frequently occur in practice, caused for example by program bugs in data preprocessing code or non-anticipated schema changes in external data sources. These errors can have devastating

Answer Sentence Selection (AS2) is an efficient approach for the design of open-domain Question Answering (QA) systems. In order to achieve low latency, traditional AS2 models score question-answer pairs individually, ignoring any information from the document each potential answer was extracted from. In contrast, more computationally expensive models designed for machine reading comprehension tasks typically

Recent advances in deep learning techniques have enabled machines to generate cohesive open-ended text when prompted with a sequence of words as context. While these models now empower many downstream applications from conversation bots to automatic storytelling, they have been shown to generate texts that exhibit social biases. To systematically study and benchmark social biases in open-ended language

Speech disfluencies are prevalent in spontaneous speech. The rising popularity of voice assistants presents a growing need to handle naturally occurring disfluencies. Semantic parsing is a key component for understanding user utterances in voice assistants, yet most semantic parsing research to date focuses on written text. In this paper, we investigate semantic parsing of disfluent speech with the ATIS

We're excited to bring the next iteration of Emora to life, putting in place new ideas and strategies that build on top of our efforts that led us to win last year’s Alexa Prize.

Our team is named in homage to Viola Spolin, originator of theater games and mother of improvisational theater.

Our goal is to advance the state of open-domain dialogue by blending modern NLP techniques with empathetic, engaging responses.

Representing SUNY Buffalo, team Proto is an amalgamation of graduate students from the Computer Science and Computational Linguistics department.

Our team is named Athena, after the Greek goddess of wisdom and strategy, because conversational AI systems need both knowledge and strategic dialogue management to be effective.

The DREAM socialbot is based on open-source platform Dp-Agent that our lab is also working on.

The first Spanish team in the history of the Alexa Challenge!

We expect Miutsu to incorporate multi-modal (voice and vision) customer experiences to allow for more efficient and intuitive interactions between customers and machines.

The GRILL team consists of School of Computing Science graduate students Carlos Gemmell, Federico Rossetto, Iain Mackie, Sophie Fischer, and Paul Owoicho with Dr. Jeff Dalton as faculty advisor.

We are a mix of graduate and undergraduate students with expertise in natural language processing, machine learning, and systems development.

This is our first time participating in any Alexa competition and we are excited to tackle the problems facing Alexa in the domains of cooking and do-it-yourself tasks.

Our ambition is to move AI and search closer to the way humans communicate and solve tasks.