talks
2025
A Cache Coherence-Aware Runtime Reconfigurable System-on-Chip for Efficient Language Model Inference
IBM AI Hardware Forum
Poster Session · November 2025 · New York, NY, USA
An Open-Source DNN Acceleration Fabric with Flexible Inter-Layer Pipelining Support
Fourth Workshop on Open-Source Computer Architecture Research (OSCAR)
Oral Presentation · June 2025 · Tokyo, Japan
Design- and Run-time Reconfigurable System-on-Chip for Language Model Inference on Edge
Columbia AI Summit
Poster Session · March 2025 · New York, NY, USA
[Slides]