BYBing Yang

Bing Yang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Bing Yang has a total of 12 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Fast and Accurate Single-Event Transient Reliability Analysis of Logic Circuits Using Adaptive Importance Sampling Method

    by Liangquan Qiao, Fangfa Fu, Guolin Ning, He Zhou, Bing Yang, Jinxiang Wang

    In nanometric integrated circuits, reliability has become one of the main design measures for space and safety-critical applications. This makes the reliability evaluation process an inevitable part of the circuit design flow. The Monte Carlo method typically adopts random sampling for single-event transient reliability evaluation, which leads to large variance in the estimated reliability. To address this issue, we develop an adaptive importance sampling (AIS) method for early stages of the design. AIS has several iterations of sampling region adjustments. By iteratively searching for failure regions, AIS may lead to better efficiency and accuracy. Additionally, a fast fault injection method is proposed to support adaptive importance sampling and multi-dimensional reliability analysis for logic circuits. Furthermore, we design an open-source netlist-level fault injection tool named NFIT . The convergence speed of the AIS method has been demonstrated to achieve up to an 18x acceleration compared to the Monte Carlo method on the ISCAS’85, EPFL, and ALS benchmark suites. We use the NFIT tool to evaluate the effect of circuit structure and targeted hardening on circuit reliability. The results indicate that, for the same Boolean function, the circuit structure has a notable influence on reliability. They also show that hardening a few critical gates can significantly improve circuit reliability.


  • Soft Error Reliability Analysis of Vision Transformers

    by Xinghua Xue, Cheng Liu, Ying Wang, Bing Yang, Tao Luo, Lei Zhang, Huawei Li, Xiaowei Li

    Vision transformers (ViTs) that leverage self-attention mechanism have shown superior performance on many classical vision tasks compared to convolutional neural networks (CNNs) and gain increasing popularity recently. Existing ViTs’ works mainly optimize performance and accuracy, but ViTs’ reliability issues induced by soft errors in large-scale VLSI designs have generally been overlooked. In this work, we mainly study the reliability of ViTs and investigate the vulnerability from different architecture granularities ranging from models, layers, modules, and patches for the first time. The investigation reveals that ViTs with the self-attention mechanism are generally more resilient on linear computing including general matrix–matrix multiplication (GEMM) and full connection (FC) and show a relatively even vulnerability distribution across the patches. ViTs involve more fragile non-linear computing such as softmax and GELU compared to typical CNNs. With the above observations, we propose a lightweight block-wise algorithm-based fault-tolerance (LB-ABFT) approach to protect the linear computing implemented with distinct sizes of GEMM and apply a range-based protection scheme to mitigate soft errors in non-linear computing. According to our experiments, the proposed fault-tolerant approaches enhance ViTs’ accuracy significantly with minor computing overhead in the presence of various soft errors.