DARTH-PUM integrates analog and Boolean PUM with optimized peripherals, coordination hardware, and a programming interface to run kernels like AES, CNNs, and LLMs fully in memory, achieving speedups of 59.4x, 14.8x, and 40.8x over an analog-plus-CPU baseline.
Nl-dpe: An analog in-memory non-linear dot product engine for efficient cnn and llm inference,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.AR 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Co-design of 14.5x compacted index, asynchronous scheduler, and multiplication-free kernel for PIM-based graph ANNS delivers up to 20x CPU and 17.1x GPU throughput on billion-scale benchmarks.
citing papers explorer
-
DARTH-PUM: A Hybrid Processing-Using-Memory Architecture
DARTH-PUM integrates analog and Boolean PUM with optimized peripherals, coordination hardware, and a programming interface to run kernels like AES, CNNs, and LLMs fully in memory, achieving speedups of 59.4x, 14.8x, and 40.8x over an analog-plus-CPU baseline.
-
Co-Designing Graph-based Approximate Nearest Neighbor Search at Billion Scale for Processing-in-Memory
Co-design of 14.5x compacted index, asynchronous scheduler, and multiplication-free kernel for PIM-based graph ANNS delivers up to 20x CPU and 17.1x GPU throughput on billion-scale benchmarks.