Yuchi Wang, Yishuo Cai, Shuhuai Ren, Sihan Yang, Linli Yao, Yuanxin Liu, Yuanxing Zhang, Pengfei Wan, Xu Sun. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Inferring the fine-grained urban flows based on the coarse-grained flow observations is practically important to many smart city-related applications. Adequate data is usually a prerequisite for existing machine learning methods, especially most deep learning models. However, many cities still suffer from the data scarcity issue due to the unbalanced city development levels. To mitigate this issue, we propose a novel cross-city fine-grained urban flow inference model named FGITrans, which aims to effectively transfer the knowledge from the data-rich cities to the data-scarce cities. Specifically, we design a weight-sharing triple-branch transformer framework which adopts self-attention and cross-attention for source/target city feature learning and domain alignment, respectively. Then, we propose a novel spatio-temporal adaptive embedding (STAE) layer for our transformer framework, and introduce a cross-city knowledge distillation (CKD) loss to narrow the cross-city disparities. The CKD loss explicitly enforces the framework to learn the discriminative domain-specific and domain-invariant representations simultaneously. Extensive experiments conducted on four large real-world datasets validate the effectiveness of FGITrans compared with the state-of-the-art baselines.

Model quantization is widely used to compress and ac-celerate deep neural networks. However, recent studies have revealed the feasibility of weaponizing model quan-tization via implanting quantization-conditioned backdoors (QCBs). These special backdoors stay dormant on released full-precision models but will come into effect after stan-dard quantization. Due to the peculiarity of QCBs, existing defenses have minor effects on reducing their threats or are even infeasible. In this paper, we conduct the first in-depth analysis of QCBs. We reveal that the activation of existing QCBs primarily stems from the nearest rounding operation and is closely related to the norms of neuron-wise truncation errors (i.e., the difference between the continuous full-precision weights and its quantized version). Motivated by these insights, we propose Error-guided Flipped Rounding with Activation Preservation (EFRAP), an effective and practical defense against QCBs. Specifically, EFRAP learns a non-nearest rounding strategy with neuron-wise er-ror norm and layer-wise activation preservation guidance, flipping the rounding strategies of neurons crucial for back-door effects but with minimal impact on clean accuracy. Ex-tensive evaluations on benchmark datasets demonstrate that our EFRAP can defeat state-of-the-art QCB attacks under various settings. Code is available here.

Model quantization is a compression technique that converts a full-precision model to a more compact low-precision version for better storage. Despite the great success of quantization, recent studies revealed the feasibility of malicious exploiting model quantization via implanting quantization-conditioned backdoors (QCBs). These special backdoors remain dormant in full-precision models but are exposed upon quantization. Unfortunately, existing defenses have limited effects on mitigating QCBs. In this paper, we conduct an in-depth analysis of QCBs. We reveal an intriguing characteristic of QCBs, where activation of backdoor-related neurons on even benign samples enjoy a distribution drift after quantization, although this drift is more significant on poisoned samples. Motivated by this finding, we propose to purify the backdoor-exposed quantized model by aligning its layer-wise activation with its full-precision version. To further exploit the more pronounced activation drifts on poisoned samples, we design an additional module to layer-wisely approximate poisoned activation distribution based on batch normalization statistics of the full-precision model. Extensive experiments are conducted, verifying the effectiveness of our defense. Our code is publicly available.

The widespread adoption of mobile smart devices has ushered in the era of Mobile Crowd Sensing (MCS), serving as an efficient method for large-scale data collection. Inherently location-sensitive, the service construction of MCS faces a crucial challenge of Location Privacy Preservation (LPP). Prior studies for LPP often necessitate a Trusted Third Party (TTP), which is not always feasible. Moreover, these privacy-preserving techniques may inadvertently obscure dishonest or malicious behaviors, leading to compromised Quality of Service (QoS). Motivated by this, we propose a Quality-enabled Bilateral Location Privacy-Preserving (Q-BLPP) service construction scheme, ensuring Bilateral LPP without TTP, while maintaining QoS. To achieve bilateral LPP, we introduce a novel BI-LBE algorithm using Bloom Indexing (BI) and Location-Based Encryption (LBE). Additionally, for high-quality recruitment, we present a Combinatorial Multi-Armed Bandit (CMAB) approach to balance exploration and exploitation. Furthermore, to ensure privacy during recruitment, worker profiles are anonymized using differential privacy. To our knowledge, our approach is the first to integrate QoS and LPP in MCS, with theoretical proofs of truthfulness and individual rationality. Simulations demonstrate that our Q-BLPP scheme strikes a favorable balance between computational efficiency, privacy security, and service quality, outperforming existing schemes.