Ashiqur Rasulhttps://rasulashiqur.github.io/Recent content on Ashiqur RasulHugo -- gohugo.ioen-usACE: Adaptive Compute-Efficient State Space Modelshttps://rasulashiqur.github.io/projects/ace/Mon, 01 Jan 0001 00:00:00 +0000https://rasulashiqur.github.io/projects/ace/Dynamic token routing for Mamba models that adapts computation to token importance.CASPER: Adaptive Retrieval with In-DRAM Processinghttps://rasulashiqur.github.io/projects/casper/Mon, 01 Jan 0001 00:00:00 +0000https://rasulashiqur.github.io/projects/casper/Confidence-guided dimension adaptivity and in-DRAM similarity search for efficient information retrieval.Adaptive Testing of Compute-in-Memory GANshttps://rasulashiqur.github.io/projects/adaptive-cim-testing/Mon, 01 Jan 0001 00:00:00 +0000https://rasulashiqur.github.io/projects/adaptive-cim-testing/Backpropagation-guided test compaction for faster testing of analog compute-in-memory systems.Energy-Efficient Domain-Specific AI Models and Agentshttps://rasulashiqur.github.io/projects/efficient-ai-agents/Mon, 01 Jan 0001 00:00:00 +0000https://rasulashiqur.github.io/projects/efficient-ai-agents/A cross-layer perspective on domain-specific intelligence, brain-inspired computing, and efficient hardware.Dynamic 3D CNN Pruning for Efficient Edge Inferencehttps://rasulashiqur.github.io/projects/dynamic-3d-cnn-pruning/Mon, 01 Jan 0001 00:00:00 +0000https://rasulashiqur.github.io/projects/dynamic-3d-cnn-pruning/Input-dependent frame, channel, and feature adaptation to reduce the cost of video inference.