Hybrid events

This thesis presents a comprehensive framework for developing an autonomous quadruped guide dog robot while aligning with the practical needs of BLV users.

This dissertation presents five systems that jointly optimize computation and communication across two workload regimes.

LLMs aid science but struggle to generalize. This talk explores capturing expert intuition and building agents for complex scientific reasoning.

This thesis investigates how to design high-throughput, cost-efficient inference serving systems that adapt to time-varying workloads.

This session event will showcase the work of the 2025–2026 PIT Faculty Fellows as they present early outcomes from their Responsible AI seed-funded projects.