Related papers: QSilk: Micrograin Stabilization and Adaptive Quant…

Recent generative-prior-based methods have shown promising blind face restoration performance. They usually project the degraded images to the latent space and then decode high-quality faces either by single-stage latent optimization or…

Recently introduced Surface Nanoscale Axial Photonics (SNAP) makes it possible to fabricate high Q-factor microresonators and other photonic microdevices by dramatically small deformation of the optical fiber surface. To become a practical…

In recent years, diffusion models have demonstrated remarkable success in high-fidelity image synthesis. However, fine-tuning these models for specialized domains, such as medical imaging, remains challenging due to limited domain-specific…

Identifying quantum flakes is crucial for scalable quantum hardware; however, automated layer classification from optical microscopy remains challenging due to substantial appearance shifts across different materials. This paper proposes a…

Silicon-based terahertz quantum cascade lasers (QCLs) offer potential advantages over existing III-V devices. Although coherent electron transport effects are known to be important in QCLs, they have never been considered in Si-based device…

Latent diffusion models have emerged as the leading approach for generating high-quality images and videos, utilizing compressed latent representations to reduce the computational burden of the diffusion process. While recent advancements…

Identifying dense subgraphs known as quasi-cliques is pivotal in numerous graph mining tasks across domains such as social networks, biology, and e-commerce. While prior work has developed efficient algorithms for quasi-clique detection in…

We introduce softpick, a rectified, not sum-to-one, drop-in replacement for softmax in transformer attention mechanisms that eliminates attention sink and massive activations. Our experiments with 340M and 1.8B parameter models demonstrate…

Manipulating fragile deformable containers, such as disposable plastic cups filled with liquid, demands real-time grip-force adaptation within an extremely narrow force margin: insufficient force causes slip, while excessive force…

Superconducting quantum circuits are a key platform for advancing quantum information processing and simulation. Scaling efforts currently encounter challenges such as Josephson-junction fabrication yield, design frequency targeting, and…

In this work, we present a highly efficient quantitative phase imaging (QPI) approach using programmable annular LED illumination based on traditional bright-field microscope. As a new type of coded illumination, the LED array provides a…

Recent advances in fast sampling methods for diffusion models have demonstrated significant potential to accelerate generation on image modalities. We apply these methods to 3-dimensional molecular conformations by building on the recently…

Quantitative susceptibility mapping (QSM) is a useful magnetic resonance imaging (MRI) technique which provides spatial distribution of magnetic susceptibility values of tissues. QSMs can be obtained by deconvolving the dipole kernel from…

Despite mounting evidence that materials imperfections are a major obstacle to practical applications of superconducting qubits, connections between microscopic material properties and qubit coherence are poorly understood. Here, we perform…

Deepfake detection has become a fundamental component of modern media forensics. Despite significant progress in detection accuracy, most existing methods remain computationally intensive and parameter-heavy, limiting their deployment on…

Diffusion transformers have recently delivered strong text-to-image generation around 1K resolution, but we show that extending them to native 4K across diverse aspect ratios exposes a tightly coupled failure mode spanning positional…

Quantum computing crucially relies on maintaining quantum coherence for the duration of a calculation. Bosonic quantum error correction protects this coherence by encoding qubits into superpositions of noise-resilient oscillator states. In…

Diffusion transformers (DiTs) achieve high generative quality but lock FLOPs to image resolution, limiting principled latency-quality trade-offs, and allocate computation uniformly across input spatial tokens, wasting resource allocation to…

Diffusion models have recently achieved remarkable performance in image super-resolution (SR), but their high computational cost limits practical deployment in remote sensing applications. To address this issue, we propose SlimDiffSR, a…

Diffusion Transformers achieve impressive generative quality but remain computationally expensive due to iterative sampling. Recently, dynamic resolution sampling has emerged as a promising acceleration technique by reducing the resolution…