SCI Doctoral Guild Student Speaker Series - October 2026

Join us for the first Doctoral Guild Student Speaker Series event for 26-27 in October 2026, with speakers Nitish Chandra and Andrew Aquilina! Catering will be provided.

SCI Doctoral Guild Student Speaker Series, Oct. 2026
Time: Oct. 16th, 12:00pm - 2:00pm
Location: Rm 538-539, 130 N. Bellefield Ave., Pittsburgh, PA, 15260

12:00pm - 12:30pm Arriving, Catering Available
12:30pm - 01:00pm Nitish's Talk (20min talk + 10min Q&A)
01:00pm - 01:30pm Andrew's Talk (20min talk + 10min Q&A)
01:30pm - 02:00pm Socializing

----------
Nitish Chandra (Home Page)

Title: The Road to Scalable Fault-Tolerant Quantum Computing

Abstract: Realizing large-scale fault-tolerant quantum computing requires overcoming fundamental challenges in hardware, quantum error correction, and reliable logical operations. Distributed quantum computing offers a promising path toward scalability by interconnecting smaller quantum processors through quantum networks, enabling computation beyond the capabilities of individual devices. However, such architectures introduce additional challenges arising from noisy interconnects, limited communication resources, and the coordination of quantum operations across multiple processors. This talk highlights recent advances in distributed quantum error correction and fault-tolerant logical operations, with an emphasis on protecting and processing quantum information reliably across interconnected quantum processors. It further discusses broader architectural challenges involving quantum network design, resource allocation, and network-aware compilation and scheduling. These research directions highlight the importance of jointly considering quantum error correction, communication, and system-level architecture in the pursuit of scalable fault-tolerant quantum computing.

Bio: Nitish Chandra is a PhD student in the Department of Informatics and Networked Systems at the University of Pittsburgh. His research focuses on distributed quantum computing, quantum error correction, and scalable fault-tolerant quantum computing architectures. His work has been published and presented at venues including IEEE Quantum Week and IEEE Transactions on Quantum Engineering. He was also a research intern at Cisco Quantum Lab, where he worked on quantum compilation and fault-tolerant quantum computing. He holds a BS-MS degree in Physics from the Indian Institute of Science Education and Research (IISER) Kolkata.

----------
Andrew Aquilina (Home Page)

Title: Lost in Delusion: Examining LLM Safety Under User Delusions and Distress

Abstract: People increasingly turn to chatbots when they're struggling, including those whose distress is tangled up with delusional beliefs, like believing the chatbot is sentient. I'll present a study where we simulated multi-turn conversations between clinically grounded personas and six LLMs, pairing each delusional conversation with a matched distress-only one. We find a recognition-intervention gap: models recognise distress about as well either way, but once it's wrapped in a delusion, some rarely act on it (up to 4.5 times fewer safety interventions). Asking models to assess distress before replying doesn't close the gap. Delusion-aware prompting with explicit guidance does narrow it, but it relies on delusion detection that's unreliable in the most vulnerable models. 

Bio: Andrew Aquilina is a third-year Ph.D. student in Information Science at the University of Pittsburgh, where he's advised by Yu-Ru Lin and is part of the Pitt NLP group. His research looks at how socially aware NLP systems assess and affect people's mental well-being: how communities and LLMs judge distress, how chatbots support users over multi-turn conversations, and when that support becomes harmful. Originally from Malta, he studied AI at the University of Malta and Stockholm University in Sweden before coming to Pitt.

Doctoral Guild at Pitt | Website | View More Events