This paper proposes a novel self-supervised clustering framework for automatic defect annotation in semiconductor manufacturing, aiming to reduce the heavy reliance on expert-labeled data. Building upon the Semantic Clustering by Adopting Nearest-neighbors (SCAN) model, we introduce a new SCAN methodology that integrates three key innovations: a redesigned pretext task, an enhanced SCAN model with neighbor loss, and a self-labeling mechanism. The new pretext task replaces the traditional ResNet18 backbone with MobileNetV2 and incorporates focal loss to address class imbalance, enabling more effective feature extraction from unlabeled wafer maps. The SCAN model is further refined through a neighbor loss function that leverages high-confidence samples to iteratively correct pseudo-labels, improving clustering accuracy. Additionally, we apply strong augmentation techniques and a confidence-based thresholding strategy to enhance model robustness. Experimental evaluations on the WM-811K and MixedWM38 datasets demonstrate that the proposed unsupervised clustering framework achieves a notable accuracy improvement (from 36.5% to 57.7% and 36.09% to 57.69%, respectively) compared to the original SCAN, with further gains up to 66.20% and 64.80% using high-confidence sample filtering. The results confirm that our approach significantly enhances classification performance in unsupervised settings and offers a practical solution for scalable, automated defect labeling in intelligent manufacturing environments.

Stochastic Boolean Satisfiability (SSAT) is a powerful representation for the concise encoding of quantified decision problems with uncertainty. While it shares commonalities with quantified Boolean formula (QBF) satisfiability and has the same PSPACE-complete complexity, SSAT solving tends to be more challenging as it involves expensive model counting, a.k.a. Sharp-SAT. To date, SSAT solvers, especially those imposing no restrictions on quantification levels, remain much lacking. In this paper, we present a new SSAT solver based on the framework of clause selection and cube distribution previously proposed for QBF solving. With model counting integrated and learning techniques strengthened, our solver is general and effective. Experimental results demonstrate the overall superiority of the proposed algorithm in both solving performance and memory usage compared to the state-of-the-art solvers on a number of benchmark formulas.

Logic regression aims to find a Boolean model involving binary covariates that predicts the response of an unknown system. It has many important applications, e.g., in data analysis and system design. In the 2019 ICCAD CAD Contest, the challenge of learning a compact circuit representing a black-box input-output pattern generator in a high dimensional Boolean space is formulated as the logic regression problem. This paper presents our winning approach to the problem based on a decision-tree reasoning procedure assisted with a template based preprocessing. Our methods outperformed other contestants in the competition in both prediction accuracy and circuit size.

The 3GPP LTE series of standards have made provisions for device-to-device (D2D) and vehicle- to-everything (V2X) communication in recent releases. The interface accommodating such communication is termed the sidelink (SL), which reuses some physical layer specifications originally defined for the uplink, such as single-carrier frequency-division multiple access (SC-FDMA), with some modifications. One modification is the use of doubly dense demodulation reference signals (DMRS) in the V2X setting to assist channel estimation. An SC-FDMA channel estimator typically starts with a least-squares (LS) estimation of the channel frequency response (CFR) at the DMRS. It then obtains CFR estimates for the data subcarriers by some time-frequency filtering. This work considers SL channel estimation techniques. The proposed method employs a novel linear minimum mean-square error (LMMSE) filtering mechanism in the frequency domain to reduce the noise in the LS CFR estimates and then, depending on the DMRS density, uses particularly designed LMMSE filtering or polynomial interpolation in the time domain to obtain the desired CFR estimates for the data subcarriers. Simulation results show that the method can yield superior performance over a wide range of channel conditions.

We design a novel fuzzy controller to dynamically adjust the packet length to protect against the interference from microwave oven over the IEEE 802.11 FHSS (frequency hopping spread spectrum) wireless LAN card. The idea of adjusting the packet length under the noise environment from the measurement results about the effects of microwave ovens over the wireless LAN card is referred by Lee, Sheu, Chen, Yu and Huaung (see APCC/OECC'99, p.817-820, 1999). Simulation results show that the designed fuzzy controller can effectively improve the transmission performance in terms of network throughput.