The necessity of explicit linguistic representations has been increasingly questioned in the era of large language models (LLMs).In this work, we revisit this issue using Universal Dependencies (UD) as a case study, examining whether and in what ways this cross-lingual syntactic framework can still benefit contemporary LLMs.We focus on a cross-lingual adversarial paraphrase identification task that is designed to foreground the role of syntactic structure in semantic interpretation across languages.Within this setting, we systematically evaluate three strategies for integrating UD into LLMs: UD-Prompt, UD-Tuning, and UD-Attention.Our experiments show that, although the magnitude of gains depends on how UD-based structural priors interact with model behavior and cross-lingual variation, UD-augmented models consistently outperform their syntax-agnostic counterparts.Across strategies, we observe average accuracy improvements of 2.67%, 8.24%, and 2.53%, respectively.These findings demonstrate that linguistic knowledge remains informative for LLMs, offering practical value in cross-lingual settings where structural alignment is challenging.

Jingyu Peng, Maolin Wang, Xiangyu Zhao, Kai Zhang, Wanyu Wang, Pengyue Jia, Qidong Liu, Ruocheng Guo, Qi Liu. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

The capacity to reconstruct the coherent characteristics of fluid flows from limited observational data holds paramount significance for applications across the physical and engineering sciences. The present study conducted steady-state experiments on a centrifugal compressor, utilizing a densely distributed array of static pressure probes to measure the actual static pressure distribution within the flow field, thereby obtaining the aerodynamic performance characteristics of the centrifugal compressor. Furthermore, the study employed randomly distributed sparse pressure sensors and numerical model data of the centrifugal compressor, applying the Gappy POD method to achieve high-fidelity reconstruction of the static pressure distribution on the impeller shroud wall. By comparing the reconstructed results with the experimental measurements, the overall reliability of the data-driven algorithm was validated, and the differences in the subtle flow field characteristics were identified. This not only deepened the understanding of the complex flow dynamics and their underlying patterns within turbo machinery, but provided important theoretical foundations and technical support for the optimization design, performance prediction, and flow control of turbo machinery.

Recently, graph-based anomaly detection (GAD) has attracted rising attention due to its effectiveness in identifying anomalies in relational and structured data. Unfortunately, the performance of most existing GAD methods suffers from the inherent structural noises of graphs induced by hidden anomalies connected with considerable benign nodes. In this work, we propose SparseGAD, a novel GAD framework that sparsifies the structures of target graphs to effectively reduce noises and collaboratively learns node representations. It then robustly detects anomalies by uncovering the underlying dependency among node pairs in terms of homophily and heterophily, two essential connection properties of GAD. Extensive experiments on real-world datasets of GAD demonstrate that the proposed framework achieves significantly better detection quality compared with the state-of-the-art methods, even when the graph is heavily attacked. Code will be available at https://github.com/KellyGong/SparseGAD.git.

Subsynchronous control interaction (SSCI) has been observed in several wind farms which consist of doubly-fed induction generators (DFIG) and are radially connected to fixed series compensated transmission line. This paper studies the SSCI characteristics of wind farm consisting of different types of generators including DFIG, permanent magnet synchronous generator (PMSG), and self-excited induction generator (SEIG). Analysis results indicate that PMSG performs little influence on SSCI, but SEIG will increase system damping on SSCI greatly. Resonant frequency also increases with the rises of SEIG capacity. Equivalent impedance based analysis reveals the mechanism of the above properties. Analysis results indicate that PMSG could be removed in SSCI analysis of wind farm consisting of different type of generators, while SEIG must be taken into consideration.