3D Gaussian Splatting (3DGS) has demonstrated impressive novel view synthesis with rapid rendering speed. Current works explore its application to geometric reconstruction to fully release the potential of 3DGS. However, a trade-off exists between the rendering and the reconstruction when representing real-world scenes, as achieving detailed surface reconstruction can often compromise the rendering quality. To address this issue, we introduce Mixed Gaussian Splatting (MixedGS) which combines 2D Gaussians for accurate surface reconstruction and 3D Gaussians for high-fidelity rendering. With anchor point initialization, 3D neural Gaussians derivation, contribution-based densification and joint supervision, our method effectively leverages the advantages of both 2D and 3D Gaussian primitives while mitigating their limitations. We demonstrate on diverse real-world scenes that our method achieves competitive rendering quality and detailed geometry reconstruction at the same time.

Multi-view surface reconstruction is essential for accurate 3D modeling and high-quality novel view synthesis. Neural implicit methods, such as NeRF, exhibit excellent rendering but struggle with precise surface extraction. While 3D Gaussian splatting (3DGS) provides efficient explicit representations, it suffers from geometric inaccuracies due to misalignment and lack of strong surface constraints, especially in real-world scenarios. These issues stem from unordered Gaussian primitives, may tend to surface drift, redundancy, and blurred boundaries. To overcome these limitations, we propose a surface-aware Gaussian aggregation framework featuring an adaptive normal alignment loss that integrates rendered, depth-based, and monocular normals to enforce robust surface orientation supervision. Additionally, our surface-guided optimization strategy aligns Gaussian primitives precisely to surfaces by exploiting combined rendered and predicted geometric information. Extensive experiments demonstrate our approach achieves state-of-the-art surface reconstruction accuracy alongside superior novel view synthesis, with ablation studies confirming the efficacy of our contributions.

Behavior prediction based on historical behavioral data have practical real-world significance. It has been applied in recommendation, predicting academic performance, etc. With the refinement of user data description, the development of new functions, and the fusion of multiple data sources, heterogeneous behavioral data which contain multiple types of behaviors become more and more common. In this paper, we aim to incorporate heterogeneous user behaviors and social influences for behavior predictions. To this end, this paper proposes a variant of Long-Short Term Memory (LSTM) which can consider context information while modeling a behavior sequence, a projection mechanism which can model multi-faceted relationships among different types of behaviors, and a multi-faceted attention mechanism which can dynamically find out informative periods from different facets. Many kinds of behavioral data belong to spatio-temporal data. An unsupervised way to construct a social behavior graph based on spatio-temporal data and to model social influences is proposed. Moreover, a residual learning-based decoder is designed to automatically construct multiple high-order cross features based on social behavior representation and other types of behavior representations. Qualitative and quantitative experiments on real-world datasets have demonstrated the effectiveness of this model.

Numerous researches on adversarial black-box attacks have proved that deep neural networks have certain insecurity. However, the current black-box attack methods still have shortages in incomplete utilization of query information. The newly proposed Simulator Attack based on meta-learning shows good performance in query-efficiency but still misses some hidden information. For this disadvantage, our research finds the usability of the feature layer output information in a simulator model for the first time. Then we propose an optimized Simulator Attack+ framework based on this discovery. By conducting experiments on the CIFAR-10 and CIFAR-100 datasets, results legibly show that Simulator Attack+ can further reduce the number of consuming queries to improve query-efficiency meanwhile maintaining attack effect. Our code is available at https://github.com/Rain117E/SimulatorAttackplus.