⌘+K

Pulse Explore Communities Researchers Journals

Home Communities Explore

September 30, 2025Open Access

A Compact, Low Power Transprecision ALU for Smart Edge Devices

Ayushi Dube Gian Singh Sarma Vrudhula

View Full Paper

Ask AI

Bookmark

Share

Key Points

TALU design achieves a 54.6x reduction in power consumption compared to existing MAC units.
Experimental results show a 2x improvement in energy efficiency while maintaining throughput for ML tasks.
The design supports multiple number formats including posit, floating point, and integer with adjustable bitwidth.
Runtime reconfiguration enables optimized performance for transprecision computing in resource-constrained environments.

No abstract yet

We only have this paper's title and publication details.

Read it at the source→

Ask about this paper

Answers are grounded in this paper and cite their sources.

Ask your own question…

Implication

This work demonstrates improved energy efficiency and area reduction in smart edge devices, highlighting the potential of transprecision computing.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Dube et al. (2025) studied this question.

synapsesocial.com/papers/68e24e6fd6d66a53c2473ec6 https://doi.org/10.48550/arxiv.2510.00333

View Full Paper

Ask AI

Bookmark

Share

Keep reading

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

1Enable Lightweight and Precision-Scalable Posit/IEEE-754 Arithmetic in RISC-V Cores for Transprecision Computing2025
2Voltage Scaled Low Power DNN Accelerator Design on Reconfigurable Platform2024 · 2 citations
3A LOW POWER DYNAMIC BITWIDTH-ADAPTIVE MULTIPLY ACCUMULATE UNIT FOR TINYML ACCELERATORS2025
4PVU: Design and Implementation of a Posit Vector Arithmetic Unit (PVU) for Enhanced Floating-Point Computing in Edge and AI Applications2025 · 1 citations
5A Near Threshold SRAM Compute-in-Memory Transformer Accelerator with Adaptive Precision Scaling for Sub mW Edge AI ApplicationsOriginal Research · 2026

T-MAC: CPU Renaissance via Table Lookup for Low-Bit LLM Deployment on Edge2024
Area and Speed-Efficient Floating-Point Arithmetic Logical Unit Implementation on FPGA2024
An efficient floating point adder for low-power devices2024