Venktesh Viswanathan Associate senior lecturer

Contact

Name and title: Venktesh ViswanathanAssociate senior lecturer

Phone: +468161797

Workplace: Department of Computer and Systems Sciences Länk till annan webbplats.

Visiting address Nodhuset, Borgarfjordsgatan 12

Postal address Institutionen för data- och systemvetenskap164 25 Kista

Research groups

AI-Enabled Digital Twins Research Group

We develop digital twins of buildings, patients and healthcare processes, and study how AI can improve their capabilities. We focus on twins that are explainable and support safe decisions, from reducing energy use to improving clinical care.

Natural Language Processing Research Group

The Natural Language Processing Research Group develops, applies and evaluates NLP methods, in particular involving large language models, across various domains. We focus on topics such as privacy, explainability, and domain adaptation.

Data Science Research Group

The Data Science Research Group focuses on core data science research, as well as on applications where data science can provide insights for decision making. We formulate novel data science problems and develop algorithmic methods and methodological workflows.

About me

I am an Associate Senior Lecturer (Biträdande lektor) in the Data Science Research group at DSV, Sotckholm University.

 

My research aims to develop robust and efficient pipelines for complex knowledge intensive tasks to assist in wide range of real-world applications like healthcare, education, research and tackling disinformation. My research his entails building robust and efficient Retrieval Augmented Generation (RAG) pipelines powered by Large Language Models (LLMs) through theoretically grounded advances in Machine Learning (ML) that bridge the retrieval and reasoning gaps. Towards realizing this vision, I have developed theoretically grounded sample-efficient algorithms for compute-optimal test time scaling at retrieval and reasoning stages. My work also involves developing robust LLM-feedback approaches based on uncertainty quantification to improve retrieval and reasoning at inference time. The resulting works have been published at WWW,WSDM,ECIR,CIKM,EMNLP, ICML, NAACL and SIGIR. I also regularly serve as Reviewer for NAACL,ACL,EMNLP,SIGKDD,SIGIR,AAAI,CIKM,WSDM,WWW,ECIR. I also serve as reviewer for journals like ACM TOIS. I am also currently serving as Program chair for Reproducibility Track at ECIR 2026.

 

Apart from publications my work has been deployed at scale for real-world impactful applications like Live-factchecking of US presidential and EU debates by Factiverse AI, Norway. My work on efficient retrieval for live fact-checking with collaborators from UIS and Factiverse won the ``Best paper award" at ECIR 2025.

I am also honoured to be a part of ELLIS - https://ellis.eu/person/venktesh-viswanathan which is "pan-European AI network of excellence which aims to secure lasting international leadership of AI made in Europe by connecting top researchers in this field and by creating a multi-centric AI research laboratory."

I am currently teaching in the following courses:

  • Research Topics in Data Science (RTDS)
  • Foundations of Data Science (FODS)

My research broadly focuses on helping people find, verify and organize information for their Complex Information Needs. For instance, doctors in multidisciplinary teams could easily look for information like ``What is the protocol for handling the virus that causes acute respiratory failure?" through such systems and save a lot of time in manual search over multiple systems for differential diagnosis to get the protocol checklist. However, the systems need to perform multi-turn information retrieval and reasoning over information to identify possible viruses causing this symptoms and protocols identified over the years to handle them and ranking them internally to synthesize the best checklist. Hence my research is at the intersection of NLP and IR also concerned with developing efficient Machine learning mechanisms for optimizing these pipelines. Departing from current view that more data and large parametric models are necessary to do this, I propose sample-efficient and compute-efficient pipelines by advancing research in:

  • Efficient neural retrieval for queries over large corpora (Millions of documents)
  • Efficient reasoning (efficient search over possible hypothesis for solutions).
  • Theoretically grounded bandit frameworks for sample-efficient exploration.
  • Promoting model reuse by evolving new skills through model merging.

 

 

P. Chungkham, V. Viswanathan, V. Setty, and A. Anand Think Right, Not More: Test-Time Scaling for Numerical Claim Verification 2025 In: Findings of the Association for Computational Linguistics: EMNLP 2025 / [ed] Christos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng, Association for Computational Linguistics , 2025, p. 24345-24363Conference paper (Refereed)

Contact

Name and title: Venktesh ViswanathanAssociate senior lecturer

Phone: +468161797

Workplace: Department of Computer and Systems Sciences Länk till annan webbplats.

Visiting address Nodhuset, Borgarfjordsgatan 12

Postal address Institutionen för data- och systemvetenskap164 25 Kista

Research groups

AI-Enabled Digital Twins Research Group

We develop digital twins of buildings, patients and healthcare processes, and study how AI can improve their capabilities. We focus on twins that are explainable and support safe decisions, from reducing energy use to improving clinical care.

Natural Language Processing Research Group

The Natural Language Processing Research Group develops, applies and evaluates NLP methods, in particular involving large language models, across various domains. We focus on topics such as privacy, explainability, and domain adaptation.

Data Science Research Group

The Data Science Research Group focuses on core data science research, as well as on applications where data science can provide insights for decision making. We formulate novel data science problems and develop algorithmic methods and methodological workflows.