About Me

I’m Mona Moghadampanah, a PhD student in Computer Science at Virginia Tech, advised by Dr. Dimitrios Nikolopoulos and a member of the PEARL research group.
My work explores systems, high-performance computing (HPC), and large language models (LLMs), with an emphasis on scaling inference across heterogeneous datacenters.
I’m passionate about building efficient, reliable AI systems and collaborating across disciplines to bridge research and real-world impact.

News

• [08/2026] Selected to serve on the Artifact Evaluation Committee for IISWC 2026.
• [06/2026] 🎉 Our paper “DiffPro: Joint Timestep and Layer-Wise Precision Optimization for Efficient Diffusion Inference” was accepted to ECCV 2026.
• [05/2026] Presented our paper “Modality Inflation: Energy Characterization and Optimization Opportunities for MLLM Inference” during the IPDPS 2026 in New Orleans, LA.
• [02/2026] 🎉 Our paper “Modality Inflation: Energy Characterization and Optimization Opportunities for MLLM Inference” was accepted to iWAPT 2026, co-located with IPDPS 2026.
• [02/2026] Selected to serve on the Artifact Evaluation Committee for ASPLOS 2026.
• [10/2025] 🎉 Selected for the TCPP/TCHPC HPCSC 25–26 Travel Award to attend SC 2025 and IPDPS 2026.
• [09/2025] 🎉 My poster “Energy-Efficient Multimodal LLM Inference: Stage-Level Characterization and Input-Aware Controls” was accepted to SC 2025, to be presented in St. Louis, MO.
• [08/2025] 🎉 Awarded a scholarship to attend the Tapia Conference 2025 in Dallas, TX.
• [08/2024] 🎓 Started my PhD in Computer Science at Virginia Tech.
• [01/2023 – 05/2024] Worked on “Memory Scavenger: A Black-Box Approach for Monitoring and Harvesting Unused Memory” — refined a cloud memory-management framework to monitor and reclaim unused memory, implemented regression-based predictors, and validated using CloudSuite and TailBench workloads under supervision of Dr. Ahmad Javadi at Amirkabir University of Technology.