BitFair: A 12-nm Bit-Serial CNN Accelerator with Learnable Early Termination and Adaptive Bit Ordering for Ultra-Low-Power XR Vision
Not on the map. Why?
Abstract (arXiv)
Extended Reality (XR) wearables require always-on perception within tight power envelopes of a few watts and motion-to-photon latency budgets below 20 ms, leaving only a few milliseconds for neural-network inference. Bit-serial computing is attractive for such energy-efficient neural network acceleration, but many existing architectures still process all bits even when ReLU sets the final output to zero. This paper presents BitFair, a software-hardware co-designed bit-serial CNN accelerator with learnable bit-level early termination and adaptive bit ordering, working under the ultra-low-power and strict latency requirements of XR applications. BitFair exploits dynamic bit-level sparsity by learning per-layer thresholds that trigger early termination when partial sums reliably predict that the final ReLU output will be zero. Furthermore, it searches for layer-wise bit orders that prioritize informative bits, maximizing early termination without sacrificing accuracy. A GlobalFoundries 12-nm FinFET implementation with a core area of 0.34 mm^2, 104 KB on-chip memory, and voltage scaling from 0.55 to 0.70 V achieves sub-millisecond latency, up to 117.0 BTOPS/W, and 0.07 pJ/SOP. On IBM DVS128 Gesture and N-MNIST, BitFair achieves 96.5% and 97.7% accuracy, respectively, while improving effective energy efficiency by 4.0-22.1x and accuracy by up to 9.2% over prior fabricated XR vision accelerators.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 18 of its 46 references that are in the map. First posted on arXiv 4 Jul 2026.
Builds on
18 of 46 references.
References · 18
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
- 2014Microsoft COCO: Common Objects in Contextcited by 17,498
- 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
- 2013Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computationcited by 2,031
- 2017Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inferencecited by 1,548
- 2019Surrogate Gradient Learning in Spiking Neural Networkscited by 604
- 2016DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradientscited by 557
- 2016Learning Structured Sparsity in Deep Neural Networkscited by 538
- 2015Converting Static Image Datasets to Spiking Neuromorphic Datasets Using Saccadescited by 308
- 2023SpikingJelly: An open-source machine learning infrastructure platform for spike-based intelligencecited by 163
- 2019Visual Wake Words Datasetcited by 46
- 2022ReckOn: A 28nm Sub-mm2 Task-Agnostic Spiking Recurrent Neural Network Processor Enabling On-Chip Learning over Second-Long Timescalescited by 35
6 more
- 2023Direct Learning-Based Deep Spiking Neural Networks: A Reviewcited by 31
- 2016Bit-pragmatic Deep Neural Network Computingcited by 30
- 2023To Spike or Not To Spike: A Digital Hardware Perspective on Deep Learning Accelerationcited by 10
- 2025BitWave: Exploiting Column-Based Bit-Level Sparsity for Deep Learning Accelerationcited by 3
- 2026SparseCol: A 1320 BTOPS/W Precision-scalable NPU Exploiting Training-free Structured Bit-level Sparsity and Dynamic Dataflowcited by 1
- 2026Mega: A 22 nm Convolutional Spiking Neural Network Accelerator Achieving 0.375 pJ/SOP for Efficient Edge Visioncited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.