arXiv subjects

Cheng Li

Publications and source records attributed to Cheng Li.

At least 127 records · Page 7Linked to original sources

arXiv2025

Can We Conjugate the Orbital Angular Momentum of a Single Photon with Nonlinear Optics? A Theoretical Analysis

Stimulated parametric down-conversion (StimPDC) produces an idler beam that exhibits a phase conjugate of the spatial structure of the classical seed beam. Extending this effect to the single-photon regime can, in principle, enable the control of the orbital angular momentum of single photons. However, its experimental implementation remains challenging due to the lower photon flux of realistic single-photon sources compared to a classical pump beam, which causes the StimPDC process to occur less frequently compared to its spontaneous counterpart. To overcome this, we propose a cascaded parametric down-conversion scheme, in which the signal photon from a spontaneously down-converted pair seeds the StimPDC process, while the idler photon heralds the seed's presence. This configuration uses coincidence detection to effectively post-select for stimulated events among the background of spontaneous contributions. We show that phase conjugation at the single-photon level manifests as a bias towards equal orbital angular momentum between the heralding idler and the stimulated idler. Our results offer a practical route to manipulate single-photon correlations in the orbital angular momentum degree of freedom and could have implications for applications such as free-space quantum communication and quantum-enhanced sensing.

physics.optics↗

arXiv2025

Lopsided and Bulging Distribution of Satellites around Paired Halos. II. 3D Analysis and Dependence on Projection and Selection Effects

We use the Illustris-TNG simulation to investigate the anisotropic distribution of subhalos in/around dark matter halo pairs. We measure the position angle ($θ$) of each subhalo by the angle between the line connecting it to the nearest host halo and the line connecting the paired halos, and examine $P(\cosθ)$ (the distribution of $\cosθ$ of all subhalos) for halo pairs with various separations ($d_{\text{sep}}$), primary halo masses ($M_p$) and secondary-to-primary halo mass ratios ($M_s/M_p$). We find that $P(\cosθ)$ generally exhibits a combined result of two distinct features: the `bulging' distribution characterized by an overabundance along the pairwise direction, and the `lopsided' distribution showing an overabundance in the region between the paired halos. The bulging signal is stronger for halo pairs with larger $d_{\text{sep}}$ and smaller $M_p$, while the lopsidedness strengthens as $M_p$ increases. Both signals depend weakly on $M_s/M_p$, and are primarily contributed by subhalos that are relatively distant from host halos. Remarkably, these measurements can be broadly reproduced by the overlap effect, provided the spatial alignment of halos is properly taken into account. Our findings suggest distinct origins: lopsidedness arises from simple halo overlap, while bulging reflects alignment with large-scale filaments. We examine the impact of projection and selection effects by conducting the same analysis in two dimensions and in a mock catalog that replicates the selection effects of the SDSS galaxy sample. We find that the 3D-to-2D projection significantly suppresses the bulging distribution, with particularly strong effects at large $d_{\text{sep}}$, small $M_p$, and large $M_s/M_p$.

astro-ph.CO↗

arXiv2025

Lopsided and Bulging Distribution of Satellites around Paired Halos. I. Observational Measurements and Comparison with Halo-based Models

We investigate the angular distribution of satellite galaxies in and around pairs of galaxy groups in SDSS and compare the results with the satellite distribution in pairs of dark matter halos by constructing mock catalogs that account for the same selection effects as the observational sample. We find that the angular distribution of satellites in both SDSS and the mock catalog exhibits a pronounced tendency towards lopsidedness, with satellites preferentially located between the two central galaxies. Additionally, there is a significant bulging distribution characterized by a higher concentration of satellites along the line connecting the two centrals compared to those found perpendicular to it. The lopsided and bulging distributions strengthen as pair separation and halo mass increase. The mock catalog successfully reproduces the observational results across all cases considered. We find that the lopsided and bulging distribution of satellites can largely be explained by overlapping two randomly selected halos matched in mass to the actual halo pairs, along with their surrounding satellite distribution, provided that the alignment between the orientations of the halos and the line connecting the halo pairs is taken into account. This suggests that the angular distribution of satellites is a natural consequence of the formation and evolution of large-scale structure in a $Λ$CDM universe, eliminating the need to introduce other physical origins.

astro-ph.GA↗

arXiv2025

Generative Inversion for Property-Targeted Materials Design: Application to Shape Memory Alloys

The design of shape memory alloys (SMAs) with high transformation temperatures and large mechanical work output remains a longstanding challenge in functional materials engineering. Here, we introduce a data-driven framework based on generative adversarial network (GAN) inversion for the inverse design of high-performance SMAs. By coupling a pretrained GAN with a property prediction model, we perform gradient-based latent space optimization to directly generate candidate alloy compositions and processing parameters that satisfy user-defined property targets. The framework is experimentally validated through the synthesis and characterization of five NiTi-based SMAs. Among them, the Ni$_{49.8}$Ti$_{26.4}$Hf$_{18.6}$Zr$_{5.2}$ alloy achieves a high transformation temperature of 404 $^\circ$C, a large mechanical work output of 9.9 J/cm$^3$, a transformation enthalpy of 43 J/g , and a thermal hysteresis of 29 °C, outperforming existing NiTi alloys. The enhanced performance is attributed to a pronounced transformation volume change and a finely dispersed of Ti$_2$Ni-type precipitates, enabled by sluggish Zr and Hf diffusion, and semi-coherent interfaces with localized strain fields. This study demonstrates that GAN inversion offers an efficient and generalizable route for the property-targeted discovery of complex alloys.

cond-mat.mtrl-sci↗

arXiv2025

Generative Artificial Intelligence in Medical Imaging: Foundations, Progress, and Clinical Translation

Generative artificial intelligence (AI) is rapidly transforming medical imaging by enabling capabilities such as data synthesis, image enhancement, modality translation, and spatiotemporal modeling. This review presents a comprehensive and forward-looking synthesis of recent advances in generative modeling including generative adversarial networks (GANs), variational autoencoders (VAEs), diffusion models, and emerging multimodal foundation architectures and evaluates their expanding roles across the clinical imaging continuum. We systematically examine how generative AI contributes to key stages of the imaging workflow, from acquisition and reconstruction to cross-modality synthesis, diagnostic support, and treatment planning. Emphasis is placed on both retrospective and prospective clinical scenarios, where generative models help address longstanding challenges such as data scarcity, standardization, and integration across modalities. To promote rigorous benchmarking and translational readiness, we propose a three-tiered evaluation framework encompassing pixel-level fidelity, feature-level realism, and task-level clinical relevance. We also identify critical obstacles to real-world deployment, including generalization under domain shift, hallucination risk, data privacy concerns, and regulatory hurdles. Finally, we explore the convergence of generative AI with large-scale foundation models, highlighting how this synergy may enable the next generation of scalable, reliable, and clinically integrated imaging systems. By charting technical progress and translational pathways, this review aims to guide future research and foster interdisciplinary collaboration at the intersection of AI, medicine, and biomedical engineering.

eess.IV↗

arXiv2025

Effective Bayesian Modeling of Large Spatiotemporal Count Data Using Autoregressive Gamma Processes

We put forward a new Bayesian modeling strategy for spatiotemporal count data that enables efficient posterior sampling. Most previous models for such data decompose logarithms of the response Poisson rates into fixed effects and spatial random effects, where the latter is typically assumed to follow a latent Gaussian process, the conditional autoregressive model, or the intrinsic conditional autoregressive model. Since log-Gaussian is not conjugate to Poisson, such implementations must resort to either approximation methods like INLA or Metropolis moves on latent states in MCMC algorithms for model fitting and exhibit several approximation and posterior sampling challenges. Instead of modeling logarithms of spatiotemporal frailties jointly as a Gaussian process, we construct a spatiotemporal autoregressive gamma process guaranteed stationary across the time dimension. We decompose latent Poisson variables to permit fully conjugate Gibbs sampling of spatiotemporal frailties and design a sparse spatial dependence structure to get a linear computational complexity that facilitates efficient posterior computation. Our model permits convenient Bayesian predictive machinery based on posterior samples that delivers satisfactory performance in predicting at new spatial locations and time intervals. We have performed extensive simulation experiments and real data analyses, which corroborated our model's accurate parameter estimation, model fitting, and out-of-sample prediction capabilities.

stat.ME↗

arXiv2025

The electroweak precision constraints of the 2HDM+S

The 2HDM+S is the singlet extension of the Two-Higgs-Doublets Model (2HDM). The singlet field and its mixing with the 2HDM Higgs sector lead to new contributions to the electroweak precision observables, in particular, the oblique parameters. In this paper, we identify five benchmark cases, where at most one mixing angle is nonzero and analyze the 95% C.L. allowed parameter space by the oblique parameters. In the alignment limit of the 2HDM, we find that other than the usual mass relations of $m_H\sim m_{H^\pm}$ or $m_A\sim m_{H^\pm}$, electroweak precision measurements also impose an upper limit on the neutral Higgs masses. In the cases with nonzero singlet mixing with the 2HDM Higgses $H$ or $A$, we find approximate mass relations of $c^2_{α_{HS}} m_{H} + s^2_{α_{HS}}m_{h_S} = m_{H^\pm}$ or $c^2_{α_{AS}} m_{A} + s^2_{α_{AS}}m_{A_S} = m_{H^\pm}$. Those relations are universal to the 2HDM+S models, with or without further symmetry assumption. We also study the non-alignment limit of the 2HDM+S, which typically has tighter constraints on the masses and mixing angles. At the end, we examine the complementarity between the electroweak precision analyses and the Higgs coupling precision measurements.

hep-ph↗

arXiv2025

KunLunBaizeRAG: Reinforcement Learning Driven Inference Performance Leap for Large Language Models

This paper introduces KunLunBaizeRAG, a reinforcement learning-driven reasoning framework designed to enhance the reasoning capabilities of large language models (LLMs) in complex multi-hop question-answering tasks. The framework addresses key limitations of traditional RAG, such as retrieval drift, information redundancy, and strategy rigidity. Key innovations include the RAG-driven Reasoning Alignment (RDRA) mechanism, the Search-Think Iterative Enhancement (STIE) mechanism, the Network-Local Intelligent Routing (NLR) mechanism, and a progressive hybrid training strategy. Experimental results demonstrate significant improvements in exact match (EM) and LLM-judged score (LJ) across four benchmarks, highlighting the framework's robustness and effectiveness in complex reasoning scenarios.

cs.AI↗

arXiv2025

Regression for Astronomical Data with Realistic Distributions, Errors and Non-linearity

We have developed a new regression technique, the maximum likelihood (ML)-based method and its variant, the KS-test based method, designed to obtain unbiased regression results from typical astronomical data. A normalizing flow model is employed to automatically estimate the unobservable intrinsic distribution of the independent variable as well as the unobservable correlation between uncertainty level and intrinsic value of both independent and dependent variables from the observed data points in a variational inference based empirical Bayes approach. By incorporating these estimated distributions, our method comprehensively accounts for the uncertainties associated with both independent and dependent variables. Our test on both mock data and real astronomical data from PHANGS-ALMA and PHANGS-JWST demonstrates that, given a sufficiently large sample size (> 1000), both the ML-based method and the KS-test based method significantly outperform the existing widely-used methods, particularly in cases of low signal-to-noise ratios. The KS-test based method exhibits remarkable robustness against deviations from underlying assumptions, complex intrinsic distributions, varying correlations between uncertainty levels and intrinsic values, inaccuracies in uncertainty estimations, outliers, and saturation effects. For sample sizes between 300 and 1000, the ML-based method yields the best performance. In the low data regime (< 300), the ML-based method maintains comparable performance to other state-of-the-art methods. A GPU-compatible Python implementation of our methods, nicknamed ``raddest'', will be made publicly available upon acceptance of this paper.

astro-ph.IM↗

arXiv2025

FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space

We present evaluation results for FLUX.1 Kontext, a generative flow matching model that unifies image generation and editing. The model generates novel output views by incorporating semantic context from text and image inputs. Using a simple sequence concatenation approach, FLUX.1 Kontext handles both local editing and generative in-context tasks within a single unified architecture. Compared to current editing models that exhibit degradation in character consistency and stability across multiple turns, we observe that FLUX.1 Kontext improved preservation of objects and characters, leading to greater robustness in iterative workflows. The model achieves competitive performance with current state-of-the-art systems while delivering significantly faster generation times, enabling interactive applications and rapid prototyping workflows. To validate these improvements, we introduce KontextBench, a comprehensive benchmark with 1026 image-prompt pairs covering five task categories: local editing, global editing, character reference, style reference and text editing. Detailed evaluations show the superior performance of FLUX.1 Kontext in terms of both single-turn quality and multi-turn consistency, setting new standards for unified image processing models.

cs.GR↗

arXiv2025

A Diffusion-Driven Temporal Super-Resolution and Spatial Consistency Enhancement Framework for 4D MRI imaging

In medical imaging, 4D MRI enables dynamic 3D visualization, yet the trade-off between spatial and temporal resolution requires prolonged scan time that can compromise temporal fidelity--especially during rapid, large-amplitude motion. Traditional approaches typically rely on registration-based interpolation to generate intermediate frames. However, these methods struggle with large deformations, resulting in misregistration, artifacts, and diminished spatial consistency. To address these challenges, we propose TSSC-Net, a novel framework that generates intermediate frames while preserving spatial consistency. To improve temporal fidelity under fast motion, our diffusion-based temporal super-resolution network generates intermediate frames using the start and end frames as key references, achieving 6x temporal super-resolution in a single inference step. Additionally, we introduce a novel tri-directional Mamba-based module that leverages long-range contextual information to effectively resolve spatial inconsistencies arising from cross-slice misalignment, thereby enhancing volumetric coherence and correcting cross-slice errors. Extensive experiments were performed on the public ACDC cardiac MRI dataset and a real-world dynamic 4D knee joint dataset. The results demonstrate that TSSC-Net can generate high-resolution dynamic MRI from fast-motion data while preserving structural fidelity and spatial consistency.

eess.IV↗

arXiv2025

NTIRE 2025 Challenge on RAW Image Restoration and Super-Resolution

This paper reviews the NTIRE 2025 RAW Image Restoration and Super-Resolution Challenge, highlighting the proposed solutions and results. New methods for RAW Restoration and Super-Resolution could be essential in modern Image Signal Processing (ISP) pipelines, however, this problem is not as explored as in the RGB domain. The goal of this challenge is two fold, (i) restore RAW images with blur and noise degradations, (ii) upscale RAW Bayer images by 2x, considering unknown noise and blur. In the challenge, a total of 230 participants registered, and 45 submitted results during thee challenge period. This report presents the current state-of-the-art in RAW Restoration.

eess.IV↗

arXiv2025

Two categories of UV-upturn galaxies revealed by semi-analytic models

UV-upturn galaxies are characterized by unusually excessive flux in the far-ultraviolet (FUV) band, observed in some elliptical galaxies and the bulges of disk galaxies. We examine UV-upturn galaxies within the semi-analytic model GABE, which embeds the formation of extreme horizontal branch (EHB) stars -- proposed as key candidates responsible for the UV-upturn phenomenon. We have analyzed all related physical processes, including stellar evolution, initial mass functions (IMFs), dust attenuation, galaxy age, metallicity, and binary fractions, in an effort to determine which processes play significant roles. Our findings reveal two categories of UV-upturn galaxies in the semi-analytic model, each with distinct formation channels: old metal-rich quenched elliptical galaxies, which are intrinsic UV-upturn galaxies induced by EHB stars within their old stellar populations, and dusty star-forming galaxies, which are relatively young and may also be photometrically identified as UV-upturn galaxies when accounting for dust attenuation. Dust attenuation contributes to 20% - 60% of the UV-upturn galaxies, depending on the specific dust attenuation models adopted. With the binary star formation model of EHB stars, both of these formation channels exhibit strong preferences for high stellar metallicity. The high-mass end slope of the IMFs is found to have a marginal effect, indicating that a universal IMF is adequate for studying the UV-upturn phenomenon.

astro-ph.GA↗

arXiv2025

Kimi k1.5: Scaling Reinforcement Learning with LLMs

Language model pretraining with next token prediction has proved effective for scaling compute but is limited to the amount of available training data. Scaling reinforcement learning (RL) unlocks a new axis for the continued improvement of artificial intelligence, with the promise that large language models (LLMs) can scale their training data by learning to explore with rewards. However, prior published work has not produced competitive results. In light of this, we report on the training practice of Kimi k1.5, our latest multi-modal LLM trained with RL, including its RL training techniques, multi-modal data recipes, and infrastructure optimization. Long context scaling and improved policy optimization methods are key ingredients of our approach, which establishes a simplistic, effective RL framework without relying on more complex techniques such as Monte Carlo tree search, value functions, and process reward models. Notably, our system achieves state-of-the-art reasoning performance across multiple benchmarks and modalities -- e.g., 77.5 on AIME, 96.2 on MATH 500, 94-th percentile on Codeforces, 74.9 on MathVista -- matching OpenAI's o1. Moreover, we present effective long2short methods that use long-CoT techniques to improve short-CoT models, yielding state-of-the-art short-CoT reasoning results -- e.g., 60.8 on AIME, 94.6 on MATH500, 47.3 on LiveCodeBench -- outperforming existing short-CoT models such as GPT-4o and Claude Sonnet 3.5 by a large margin (up to +550%).

cs.AI↗

arXiv2025

HATA: Trainable and Hardware-Efficient Hash-Aware Top-k Attention for Scalable Large Model Inference

Large Language Models (LLMs) have emerged as a pivotal research area, yet the attention module remains a critical bottleneck in LLM inference, even with techniques like KVCache to mitigate redundant computations. While various top-$k$ attention mechanisms have been proposed to accelerate LLM inference by exploiting the inherent sparsity of attention, they often struggled to strike a balance between efficiency and accuracy. In this paper, we introduce HATA (Hash-Aware Top-$k$ Attention), a novel approach that systematically integrates low-overhead learning-to-hash techniques into the Top-$k$ attention process. Different from the existing top-k attention methods which are devoted to seeking an absolute estimation of qk score, typically with a great cost, HATA maps queries and keys into binary hash codes, and acquires the relative qk score order with a quite low cost, which is sufficient for realizing top-k attention. Extensive experiments demonstrate that HATA achieves up to 7.2$\times$ speedup compared to vanilla full attention while maintaining model accuracy. In addition, HATA outperforms the state-of-the-art top-$k$ attention methods in both accuracy and efficiency across multiple mainstream LLM models and diverse tasks. HATA is open source at https://github.com/gpzlx1/HATA.

cs.LG↗

arXiv2025

Fast, Secure, Adaptable: LionsOS Design, Implementation and Performance

We present LionsOS, an operating system for security- and safety-critical embedded systems. LionsOS is based on the formally verified seL4 microkernel and designed with verification in mind. It uses a static architecture and features a highly modular design driven by strict separa- tion of concerns and a focus on simplicity. We demonstrate that LionsOS achieves excellent performance on system-call intensive workloads.

cs.OS↗

arXiv2025

Cross-Sequence Semi-Supervised Learning for Multi-Parametric MRI-Based Visual Pathway Delineation

Accurately delineating the visual pathway (VP) is crucial for understanding the human visual system and diagnosing related disorders. Exploring multi-parametric MR imaging data has been identified as an important way to delineate VP. However, due to the complex cross-sequence relationships, existing methods cannot effectively model the complementary information from different MRI sequences. In addition, these existing methods heavily rely on large training data with labels, which is labor-intensive and time-consuming to obtain. In this work, we propose a novel semi-supervised multi-parametric feature decomposition framework for VP delineation. Specifically, a correlation-constrained feature decomposition (CFD) is designed to handle the complex cross-sequence relationships by capturing the unique characteristics of each MRI sequence and easing the multi-parametric information fusion process. Furthermore, a consistency-based sample enhancement (CSE) module is developed to address the limited labeled data issue, by generating and promoting meaningful edge information from unlabeled data. We validate our framework using two public datasets, and one in-house Multi-Shell Diffusion MRI (MDM) dataset. Experimental results demonstrate the superiority of our approach in terms of delineation performance when compared to seven state-of-the-art approaches.

cs.CV↗

arXiv2025

DynaServe: Unified and Elastic Execution for Dynamic Disaggregated LLM Serving

LLM inference must meet strict latency SLOs (e.g., 100 ms P99 time-between-tokens) while maximizing goodput. Yet, real-world variability in prompt and response lengths skews compute-intensive prefill and memory-bound decode phases, making both colocated (even with chunked prefill) and disaggregated deployments unable to simultaneously deliver low tail latency and high throughput. We introduce DynaServe, a high-performance LLM serving system built atop vLLM that unifies and extends both paradigms for maximizing goodput under SLO constraints, when handling unbalanced and dynamic workloads. It relies on a micro-request abstraction, which arbitrarily splits each request at any token boundary into at most two cooperating segments. A two-level scheduling framework then balances micro-request load across unified GPU instances. The global scheduler rapidly selects per-request split points by considering both the request's prefill/decode time ratio and the current load across GPU instances. The local schedulers on each GPU instance independently form SLO-aware batches, adjusting their composition in response to workload fluctuations, potential latency spikes and per-GPU under/over utilization. On real-world traces, DynaServe boosts the overall serving capacity from 1.15$\times$ to 3.07$\times$, improves goodput by up to 1.91$\times$ and 1.61$\times$, and improves the performance by up to 60\% in a hybrid workload under SLO compared to state-of-the-art colocated and disaggregated baselines.

cs.DC↗