talks

Workshop: Teaching Neural Machine Translation to Translators

The workshop explored questions on the lines of how should knowledge about neural machine translation be disseminated in the translator/interpreter community, what is the preferred approach, and how detailed should it be. The presented is just one of …

Cognitive Natural Language Processing

In this talk, I discuss the interdisciplinary area cognitive NLP. I discuss the various intersections which give it a unique blend and discuss the basics of research based on eye-tracking and NLP. Further, I try to illustrate the nitty-gritties of …

Quality Estimation

In this talk, I discuss the Quality Estimation (QE) for Machine Translation (MT). I discuss the basics of the QE task, and the QE shared task organized every year. Further, the discussion get into how current MT systems achieve very good results on a …

Dialouge Processing: The role of NLP and building blocks of a chatbot

In this invited talk to the [LITHME WG8](https://lithme.eu/working-groups/wg8/) meeting, I gave an introduction to chatbots and its basic building blocks. We discussed the role of fundamental NLP tasks in chatbots and how they can be used to build a …

Transformers: Perspectives from NLP

In this talk, I discuss the Transformers architecture in some detail along with the evolution of embeddings in the NLP sub-area of AI. The talk also features various other architectures and example of the fine tuning process. Some slides have been …

Introducing Neural Machine Translation (NMT) to Translators

Should translators understand the power of the technology behind Google Translate and other providers of machine translation? We think so, and that is why we prepared this webinar as the first step in helping translators make informed decisions when …

Quality Estimation for Machine Translation

Current Machine Translation (MT) systems achieve very good results on a growing variety of language pairs and datasets. However, they are known to produce fluent translation outputs that can contain important meaning errors, thus undermining their …

Investigations into the use of Distributed Semantics for Cognate Detection and Computational Phylogenetics

In this talk, I present my primary PhD contributions where different embeddings aid in the tasks of cognate detection, and computational phylogenetics. A set of unpublished results which show that cognates, indeed, help in the downstream task of …

Natural Language Processing and its intersection with Data Science

In this talk, I discuss Natural Language Processing, Data Science and how they are related to each other. While presenting the different facets of the NLP research done at our lab, I connected it to how a data scientist can use the research to solve …