Self-Admitted Technical Debt (SATD) refers to sub-optimal solutions deliberately introduced to accelerate the software development process, often at the expense of software maintainability and sustainability. Therefore, timely identification and repayment of the SATD is critical for the software system. As exploration deepens, it is found that effectively prioritizing the repayment of SATD with more significant impacts on software quality requires not only identifying SATD but also further classifying it. However, existing SATD identification and classification approaches face the following challenges: (1) SATDs originate from diverse sources. Code comments are a widespread source, but recent research has revealed that SATDs can originate from other sources, such as pull requests, issues, and commit messages. Nonetheless, existing approaches primarily target code comments, lacking the capability to analyze SATDs from other sources effectively. (2) SATDs fall into diverse categories. Nonetheless, existing SATD classification approaches fail to address all SATD categories comprehensively and show inadequate performance. (3) Imbalance of existing SATD datasets. Real-world SATD data are scarce, making dataset collection challenging. Moreover, SATD distribution across different sources is uneven, further complicating the construction of high-quality datasets. To alleviate these challenges, this article presents an SATD identification and classification framework named IMPACT . First, IMPACT employs ChatGPT to construct an augmented dataset. Subsequently, it utilizes a pipeline with two fine-tuned language models of different parameter sizes to identify and classify SATD separately. To evaluate the effectiveness of IMPACT, we compare it with three state-of-the-art SATD classification methods and its two foundation models. Experimental results demonstrate that IMPACT outperforms state-of-the-art methods by a large margin, and even surpasses its foundation model GLM-4-9B-Chat. It achieves the optimal average F1 score of 0.697 on the source of pull requests, the most challenging data source. Moreover, experiments on the cross-project test set show that IMPACT demonstrates strong generalizability on unseen project data.

Time-domain signals are typically transformed using a 2D-FFT to generate the Range Doppler Map (RDM) in FMCW radar data processing. Subsequently, 2D-CFAR is applied to the RDM for target detection, while Direction of Arrival (DOA) estimation is utilized to obtain the angular information of the target, thus enabling the acquisition of the target’s spatial information. As the demand for accurate target information increases, the volume of data in each RDM frame also increases. This escalation necessitates significant hardware resources and an area to store the RDM in on-chip SRAM or a large bandwidth to accommodate it in off-chip DDRs. Furthermore, DDR controllers require substantial on-chip resources. To address this challenge, we propose a compression algorithm named APFG, which comprises Amplitude-Phase Transformation (APT), Fix-to-Float (Fix2Float), and Exponential Sharing (ESH). Experimental results demonstrate that this algorithm is applicable across various scenarios, CFAR types, and CFAR configurations, achieving high compression ratios while ensuring the accuracy of CFAR and DOA estimates. Specifically, it achieves a compression ratio of less than 14% when Pceis below 1% and a compression ratio of nearly 30% when Pdeis around 2%.

Zhiwei Fei, Xiaoyu Shen, Dawei Zhu, Fengzhe Zhou, Zhuo Han, Alan Huang, Songyang Zhang, Kai Chen, Zhixin Yin, Zongwen Shen, Jidong Ge, Vincent Ng. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.