Junkai Chen, Huihui Huang, Yunbo Lyu, Junwen An, Jieke Shi, Chengran Yang, Ting Zhang, Haoye Tian, Yikun Li, Zhenhao Li, Xin Zhou, Xing Hu, David Lo. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Coordinated cyber-physical attacks (CCPAs) pose a critical threat to the secure operation of smart power grids. While existing studies often assume simultaneous or sequence-agnostic attack strategies, this paper proposes a sequence-aware CCPA framework that explicitly models the temporal coupling between cyber manipulation and physical sabotage. We enhance the classical load redistribution attack (LRA) by addressing four key limitations: detectability due to infeasible power flows, violation of power balance, omission of post-attack system response, and insensitivity to attack sequencing. Specifically, we formulate two distinct bilevel attack models, namely Cyber-to-Physical (C→P) and Physical-to-Cyber (P→C), and solve them via an exact KKT-based Mixed-Integer Linear Programming (MILP) reformulation and a scalable Benders decomposition (BD) framework. Experiments on IEEE 14-, 57-, and 118-bus systems demonstrate that C→P attacks induce significantly more line overloads than P→C, validating the heightened risk of cyber-initiated cascades. Moreover, our BD approach accurately identifies spatial vulnerability hotspots with high fidelity, even when the physical attack budget is extended fromRp= 1 toRp= 2, confirming the framework’s scalability and practical relevance. The results provide actionable insights for adaptive grid protection against sophisticated, sequential threats.

While third-party libraries (TPLs) are extensively reused to enhance productivity during software development, they can also introduce potential security risks such as vulnerability propagation. Software composition analysis (SCA), proposed to identify reused TPLs for reducing such risks, has become an essential procedure within modern DevSecOps. As one of the mainstream SCA techniques, binary-to-source SCA identifies the third-party source projects contained in binary files via binary source code matching, which is a major challenge in reverse engineering since binary and source code exhibit substantial disparities after compilation. The existing binary-to-source SCA techniques leverage basic syntactic features that suffer from redundancy and lack robustness in the large-scale TPL dataset, leading to inevitable false positives and compromised recall. To mitigate these limitations, we introduce BinaryAI, a novel binary-to-source SCA technique with two-phase binary source code matching to capture both syntactic and semantic code features. First, BinaryAI trains a transformer-based model to produce function-level embeddings and obtain similar source functions for each binary function accordingly. Then by applying the link-time locality to facilitate function matching, BinaryAI detects the reused TPLs based on the ratio of matched source functions. Our experimental results demonstrate the superior performance of BinaryAI in terms of binary source code matching and the downstream SCA task. Specifically, our embedding model outperforms the state-of-the-art model CodeCMR, i.e., achieving 22.54% recall@1 and 0.34 MRR compared with 10.75% and 0.17 respectively. Additionally, BinaryAI outperforms all existing binary-to-source SCA tools in TPL detection, increasing the precision from 73.36% to 85.84% and recall from 59.81% to 64.98% compared with the well-recognized commercial SCA product Black Duck.

Content-based image retrieval (CBIR) aims at querying targets by using visual contents of image itself. Although classical wavelet transform is effective in representing image feature and thus is suitable in CBIR, it still encounters problems especially in implementation, e.g. floating-point operation and decomposition speed, which may nicely be solved by lifting scheme, a novel spatial approach for constructing biorthogonal wavelet filters. Lifting scheme has such intriguing properties as convenient construction, simple structure, integer-to-integer transform, low computational complexity as well as flexible adaptivity, revealing its potentials in CBIR. In this paper, by using general lifting and its adaptive version, we decompose HSI color images into multilevel scale and wavelet coefficients, with which, we can perform image feature extraction and similarity match by means of F-norm theory. Meanwhile, we provide a progressive image filtering strategy to achieve flexible CBIR. Eventually, the retrieval performances of lifting scheme are compared with those of its classical counterpart in retrieval accuracy and speed.