Arpita Saggar

Hello!

I am a PhD student within the UKRI CDT in AI for Medical Diagnosis and Care at the University of Leeds. My doctoral research focuses on using generative AI to create patient simulations for medical training. I am fortunate to be supervised by Professor David Hogg, Dr Duygu Sarikaya, Dr Jonathan Darling and Professor Vania Dimitrova.

Before starting my PhD, I was a Data Science for Social Good fellow at the University of Warwick, where I worked with Save The Children UK and UNICEF to map child poverty across sub-Saharan Africa. My background is in Computing, and I’ve previously worked on research projects at the intersection of AI and healthcare.

News

[Sep 2025] I will be presenting my poster on persona-conditioned dialogue generation at the UK AI Research Symposium in Newcastle
[Aug 2025] I will be presenting my research on ChatGPT-simulated clinical training scenarios at the AMEE Conference in Barcelona
[Jul 2025] Score Before You Speak has been accepted to the European Conference on Artificial Intelligence (ECAI)
[Jun 2025] I shared the Best Poster Prize with Yushi Guo at the Annual Conference of the AI-Medical CDT
[May 2025] I won People’s Choice Prize at the London Hopper Colloquium, organised by UCL Computer Science, EPSRC, BCS Academy of Computing and GSA Capital
[May 2025] I presented my research on ChatGPT-simulated clinical training scenarios at the Research Presentation Day organised by the British Federation of Women Graduates
[Mar 2025] My team won first place in the Easy Read Image Generation Hackathon organised by Leeds Digital Health

Research

Score Before You Speak: Improving Persona Consistency in Dialogue Generation using Response Quality Scores

Arpita Saggar, Jonathan C. Darling, Vania Dimitrova, Duygu Sarikaya, David C. Hogg

European Conference on Artificial Intelligence (2025)

Score-conditioned next token prediction to improve persona consistency in dialogue

Project Page

Paper

Code

Micro-projections of multidimensional child poverty in sub-Saharan Africa

Marina Vicini, John Fitzgerald, Daniela Pinto Veizaga, Arpita Saggar, Oliver Fiala

Handbook on Child Poverty and Inequality (2025)

Chapter on machine learning methods for producing fine-grained multidimensional child poverty estimates.

Handbook Chapter

Code

XAI-MethylMarker: Explainable AI approach for biomarker discovery for breast cancer subtype classification using methylation data

Sheetal Rajpal, Ankit Rajpal, Arpita Saggar, Ashok K Vaid, Virendra Kumar, Manoj Agarwal, Naveen Kumar

Expert Systems with Applications (2023)

Explainable AI-based biomarker discovery framework applied to DNA methylation data to determine a small biomarker set for breast cancer stratification.

Paper

Code

Patent

COV-ELM classifier: An extreme learning machine based identification of COVID-19 using chest X-ray images

Sheetal Rajpal, Manoj Agarwal, Ankit Rajpal, Navin Lakhyani, Arpita Saggar, Naveen Kumar

Intelligent Decision Technologies (2022)

A three-stage ELM-based framework to distinguish between COVID-19, normal, and pneumonia chest X-rays.

Paper

Other Activities

I am involved in relevant activites beyond my research. I’m a student representative for my CDT and helped organise our annual CDT conference in 2025 and 2024 (also moderated the industry panel in 2024). I’ve also been involved in STEM engagement through the I’m a Scientist, Get me out of here programme. During my master’s, I edited the university’s annual tech magazine, Srijan, and volunteered as a programming mentor at the Uplift Project by GirlScript Foundation.

Outside of academic life, I enjoy listening to music, swimming, playing chess and badminton, roller skating and bowling (though I’m not very good at it 😅). Below are some pictures taken during my PhD journey.