Qniverse（Quantitative Trading’s Universe）是一个基于前沿人工智能和机器学习技术的开源量化交易算法项目。项目涵盖从传统方法到深度神经网络等多种技术，致力于构建一个统一框架，为不同类型的量化算法提供公平、透明、开放的回测与评估系统。同时，项目通过适配现有的量化交易开源库，方便广大开发者进行灵活使用和高效测试。

~ Continue reading ~

We propose DNF, a dictionary-based representation for the unconditional generation of 4D deforming shapes, with a transformer-based diffusion model. Our method is capable of generating motions with superior shape quality and temporal consistency.

We propose a Locally constrained Compact point cloud Model (LCM) consisting of a locally constrained compact encoder and a locally constrained Mamba-based decoder. Our encoder replaces self-attention with our local aggregation layers to achieve an elegant balance between performance and efficiency. Considering the varying information density between masked and unmasked patches in the decoder inputs of MPM, we introduce a locally constrained Mamba-based decoder. This decoder ensures linear complexity while maximizing the perception of point cloud geometry information from unmasked patches with higher information density. Extensive experimental results show that our compact model significantly surpasses existing Transformer-based models in both performance and efficiency, especially our LCM-based Point-MAE model, compared to the Transformer-based model, achieved an improvement of 1.84%, 0.67%, and 0.60% in average accuracy on the three variants of ScanObjectNN while reducing parameters by 88% and computation by 73%.

In this paper, we introduce a novel Dual-domain Dynamic Normalization (DDN) framework designed to address the challenge of dynamically capturing distribution variations across time and frequency domains. DDN operates in a sliding window fashion, enabling it to detect subtle, time-varying changes in data distributions. DDN performs time-domain normalization to compute local sliding statistics (mean and standard deviation) at each time step, offering a fine-grained approach compared to traditional methods that operate at a coarser level.

We introduce a diffusion-based generative model that learns from just one example. Our approach involves two core components: 1) an efficient yet expressive level representation, and 2) a latent denoising network with constrained receptive fields.

In this project, we introduce a novel approach GladCoder to generate stylized QR codes that are personalized, natural, and text-driven. Its pipeline includes a Depth-guided Aesthetic QR code Generator (DAG) to improve quality of image foreground, and a GrayscaLe-Aware Denoising (GLAD) process to enhance scanning-robustness. The overall pipeline is based on diffusion models, which allow users to create stylized QR images from a textual prompt to describe the image and a textual input to be encoded.

Quantitatively, compared with Transformer-based models, PDF (720) yields an overall 14.59% reduction in MSE and 10.77% reduction in MAE. Compared with CNN-based models, PDF (720) yields an overall 24.61% reduction in MSE and 19.91% reduction in MAE. Compared with Linear-based models, PDF (720) yields an overall 7.05% reduction in MSE and 5.51% reduction in MAE.

We propose a novel Cross-Modal LLM Fine-Tuning (CALF) framework for MTSF by reducing the distribution discrepancy between textual and temporal data, which mainly consists of the temporal target branch with temporal input and the textual source branch with aligned textual input. To reduce the distribution discrepancy, we develop the cross-modal match module to first align cross-modal input distributions. Additionally, to minimize the modality distribution gap in both feature and output spaces, feature regularization loss is developed to align the intermediate features between the two branches for better weight updates, while output consistency loss is introduced to allow the output representations of both branches to correspond effectively.

In this project, we propose DiffAD, a method for unsupervised anomaly detection based on the latent diffusion model, inspired by its ability to generate high-quality and diverse images. We further propose noisy condition embedding and interpolated channels to address the aforementioned challenges in the general reconstruction-based pipeline. Extensive experiments show that our method achieves state-of-the-art performance on the challenging MVTec dataset, especially in localization accuracy.