Dr. Kuldeep


Computer Scientist  - Leader - Mentor- Speaker

About Me


I am a computer scientist by education, Sr. Product Manager by profession. I now work at Cerence (formerly Nuance Automotive) in Aachen, Germany. I am part of the In-car application team and driving innovation in conversational AI-related products for tier-1 OEMs. I am responsible for the success of a few next-gen innovative conversational AI products of Cerence. Some of the existing products I own are already on the road in leading car makers. My role is quite unconventional, and I work at the intersection of conversational AI research and business unit, with a concrete understanding of both domains. Hence, I not only understand how AI can improve businesses, but I do understand how to build AI models from scratch.

Previously I was with Fraunhofer IAIS, Germany as Senior Dialogue System Researcher where I focussed on the research domain: Dialogue Systems and Question Answering. I was leading a team that improved question-answering capabilities in Fraunhofer’s spin-off dialogue system technology Sophia. In Fraunhofer, I was working with Prof. Dr. Jens Lehmann. Prior to that, I did my Ph.D. under the supervision of Prof. Dr. Sören Auer from the University of Bonn, Germany in Question answering over Knowledge Graphs. During my Ph.D. studies, I was associated with Fraunhofer IAIS as an early-stage researcher. I received the prestigious Marie Curie Fellowship for Ph.D. studies. 

I have (co)-authored over 25 peer-reviewed publications in top conferences such as The Web Conference, NAACL, CIKM, SIGIR, ISWC, ESWC, etc. In my free time, I do open-source research in the area of semantic web and natural language processing (NLP). Specifically, I focus on Knowledge graphs and downstream NLP tasks such as entity linking, relation extraction, and question answering. For a complete list of publications, refer to my Google Scholar. I also founded a  voluntary research NGO "Zerotha" that helps students who seek guidance in research as their first step. I call it "Saturday research" and is one of my hobbies.


You can reach out to me on my email: kskuldeepvit (AT) Gmail (DOT) com 


Following are my other digital footprints: 

  • LinkedIn
  • Twitter
  • Blogger
Marble Surface
Zerotha Research

Zeroth step towards research

Image by Gijs Coolen

We are a group of independent researchers who support students seeking the first step towards their research careers.

Marble Surface


No student is good or bad, its just a matter of time, hard work, and research ecosystem you are part of. We aim to provide a research ecosystem to aspiring students who somehow are left behind.

  • May 2021

    • Paper accepted in International Conference on Web Engineering (ICWE) 2021​

      • ​Better Call the Plumber: Orchestrating Dynamic Information Extraction Pipelines.

  • March 2021

    • Two papers accepted in Extended Semantic Web Conference (ESCW) 2021​

      • ​ParaQA: A Question Answering Dataset with Paraphrase Responses for Single-Turn Conversation

      • Context Transformer with Stacked Pointer Networks for Conversational Question Answering over Knowledge Graphs

  • January 2021

    • Two papers accepted in EACL 2021​

      • CHOLAN:  A Modular Approach for Neural Entity Linking on Wikipedia and Wikidata​

      • Conversational Question Answering over Knowledge Graphs with Transformer and Graph Attention Networks

    • Paper accepted in The Web Conference 2021​

      • ​RECON: Relation Extraction using Knowledge Graph Context in a Graph Neural Network

  • October 2019

    • Arjun paper was the best paper runner up in Web Information System Engineering Conference 2020. 

  • July 2020

    • Paper "Evaluating the Impact of Knowledge Graph Context on Entity Disambiguation Models" is accepted in CIKM 2020. Preprint at: Link.

    • Arjun paper "Encoding Knowledge Graph Entity Aliases in Attentive Neural Networks for Wikidata Entity Linking" is accepted in 21st Web Information System Engineering (WISE 2020). Preprint

    • Paper accepted in the Journal of Web Semantics. A preprint is at Link.

    • Falcon 2.0: An Entity and Relation Linking Tool over Wikidata is accepted in CIKM 2020. Preprint