Search

The rapid advancement of generative artificial intelligence (AI) in recent years has profoundly reshaped modern lifestyles, necessitating a revolutionary architecture to support the growing demands for computational power. Cloud computing has become the driving force behind this transformation. However, it consumes significant power and faces computation security risks due to the reliance on extensive data centers and servers in the cloud. Reducing power consumption while enhancing computational scale remains persistent challenges in cloud computing. Here, we propose and experimentally demonstrate an optical cloud computing system that can be seamlessly deployed across edge-metro network. By modulating inputs and models into light, a wide range of edge nodes can directly access the optical computing center via the edge-metro network. The experimental validations show an energy efficiency of 118.6 mW/TOPs (tera operations per second), reducing energy consumption by two orders of magnitude compared to traditional electronic-based cloud computing solutions. Furthermore, it is experimentally validated that this architecture can perform various complex generative AI models through parallel computing to achieve image generation tasks.

The escalating demands of compute-intensive applications urgently necessitate the adoption of optical interconnect technologies to overcome bottlenecks in scaling computing systems. This requires fully exploiting the inherent parallelism of light across scalable dimensions for data loading. Here we experimentally demonstrate a synergy of wavelength- and mode- multiplexing combined with high-order modulation formats to achieve multi-tens-of-terabits-per-second optical interconnects using foundry-compatible silicon photonic circuits. Implementing an edge-guided analog-and-digital optimization method that integrates high efficiency with fabrication robustness, we achieve the inverse design of mode multiplexers based on digital metamaterial waveguides. Furthermore, we employ a packaged five-mode multiplexing chip, achieving a single-wavelength interconnect capacity of 1.62 Tbit s-1 and a record-setting multi-dimensional interconnect capacity of 38.2 Tbit s-1 across 5 modes and 88 wavelength channels, with high-order formats up to 8-ary pulse-amplitude-modulation (PAM). This study highlights the transformative potential of optical interconnect technologies to surmount the constraints of electronic links, thus setting the stage for next-generation datacenter and optical compute interconnects.

Test-time adaptation (TTA) aims to improve the performance of source-domain pre-trained models on previously unseen, shifted target domains. Traditional TTA methods primarily adapt model weights based on target data streams, making model performance sensitive to the amount and order of target data. The recently proposed diffusion-driven TTA methods mitigate this by adapting model inputs instead of weights, where an unconditional diffusion model, trained on the source domain, transforms target-domain data into a synthetic domain that is expected to approximate the source domain. However, in this paper, we reveal that although the synthetic data in diffusion-driven TTA seems indistinguishable from the source data, it is unaligned with, or even markedly different from the latter for deep networks. To address this issue, we propose a \textbf{S}ynthetic-\textbf{D}omain \textbf{A}lignment (SDA) framework. Our key insight is to fine-tune the source model with synthetic data to ensure better alignment. Specifically, we first employ a conditional diffusion model to generate labeled samples, creating a synthetic dataset. Subsequently, we use the aforementioned unconditional diffusion model to add noise to and denoise each sample before fine-tuning. This Mix of Diffusion (MoD) process mitigates the potential domain misalignment between the conditional and unconditional models. Extensive experiments across classifiers, segmenters, and multimodal large language models (MLLMs, \eg, LLaVA) demonstrate that SDA achieves superior domain alignment and consistently outperforms existing diffusion-driven TTA methods. Our code is available at https://github.com/SHI-Labs/Diffusion-Driven-Test-Time-Adaptation-via-Synthetic-Domain-Alignment., Comment: GitHub: https://github.com/SHI-Labs/Diffusion-Driven-Test-Time-Adaptation-via-Synthetic-Domain-Alignment

Two-dimensional (2D) materials provide an ideal platform to explore novel superconducting behavior including Ising superconductivity, topological superconductivity and Majorana bound states in different 2D stoichiometric Ta-, Nb-, and Fe-based crystals. However, tuning the element content in 2D compounds for regulating their superconductivity has not been realized. In this work, we report the synthesis of high quality Fe1+yTe with tunable Fe content by chemical vapor deposition (CVD). The quality and composition of Fe1+yTe are characterized by Raman spectroscopy, X-ray photoelectron spectroscopy (XPS) and scanning transmission electron microscopy (STEM). The superconducting behavior of Fe1+yTe crystals with varying Fe contents is observed. The superconducting transition of selected Fe1.13Te sample is sharp ({\Delta}Tc = 1 K), while Fe1.43Te with a high-Fe content shows a relative broad superconducting transition ({\Delta}Tc = 2.6 K) at zero magnetic field. Significantly, the conspicuous vortex flow and a transition from a 3D vortex liquid state to a 2D vortex liquid state is observed in Fe1.43Te sample. Our work highlights the tunability of the superconducting properties of Fe1+yTe and sheds light on the vortex dynamics in Fe-based superconductors, which facilitates us to understand the intrinsic mechanisms of high-temperature superconductivity., Comment: 30 pages, 15 figures, accepted by Advanced Functional Materials

At present, face 3D reconstruction has broad application prospects in various fields, but the research on it is still in the development stage. In this paper, we hope to achieve better face 3D reconstruction quality by combining multi-view training framework with face parametric model Flame, propose a multi-view training and testing model MFNet (Multi-view Flame Network). We build a self-supervised training framework and implement constraints such as multi-view optical flow loss function and face landmark loss, and finally obtain a complete MFNet. We propose innovative implementations of multi-view optical flow loss and the covisible mask. We test our model on AFLW and facescape datasets and also take pictures of our faces to reconstruct 3D faces while simulating actual scenarios as much as possible, which achieves good results. Our work mainly addresses the problem of combining parametric models of faces with multi-view face 3D reconstruction and explores the implementation of a Flame based multi-view training and testing framework for contributing to the field of face 3D reconstruction.

With the rise of short video platforms as prominent channels for news dissemination, major platforms in China have gradually evolved into fertile grounds for the proliferation of fake news. However, distinguishing short video rumors poses a significant challenge due to the substantial amount of information and shared features among videos, resulting in homogeneity. To address the dissemination of short video rumors effectively, our research group proposes a methodology encompassing multimodal feature fusion and the integration of external knowledge, considering the merits and drawbacks of each algorithm. The proposed detection approach entails the following steps: (1) creation of a comprehensive dataset comprising multiple features extracted from short videos; (2) development of a multimodal rumor detection model: first, we employ the Temporal Segment Networks (TSN) video coding model to extract video features, followed by the utilization of Optical Character Recognition (OCR) and Automatic Speech Recognition (ASR) to extract textual features. Subsequently, the BERT model is employed to fuse textual and video features; (3) distinction is achieved through contrast learning: we acquire external knowledge by crawling relevant sources and leverage a vector database to incorporate this knowledge into the classification output. Our research process is driven by practical considerations, and the knowledge derived from this study will hold significant value in practical scenarios, such as short video rumor identification and the management of social opinions.