arXiv subjects

Xuan Yang

Publications and source records attributed to Xuan Yang.

At least 55 records · Page 3Linked to original sources

arXiv2025

High-pulse-energy integrated mode-locked lasers based on a Mamyshev oscillator

Ultrafast lasers have unlocked numerous advances across science and technology: they enable corneal surgery, reveal chemical reaction dynamics, and underpin optical atomic clocks. Over the past decades, extensive efforts have been devoted to developing photonic integrated circuit-based mode-locked lasers that are compact, scalable, and compatible with further on-chip functionalities. Yet, existing implementations fall short of pulse energies required for their subsequent uses in nonlinear applications. In this work, we demonstrate the first mode-locked laser that overcomes this limitation in low-loss erbium-doped silicon nitride photonic integrated circuits. The laser is based on the Mamyshev oscillator architecture, which employs alternating spectral filtering and self-phase modulation for mode-locking. It delivers a 176 MHz stream of pulses with nanojoule energy, comparable to fiber lasers and surpassing previous photonic integrated sources by more than two orders of magnitude. The output pulses exhibit excellent coherence, can be linearly compressed to 147 fs and directly drive a 1.5-octave-spanning supercontinuum in an integrated waveguide. Our work establishes a new generation of high-pulse-energy photonic integrated mode-locked lasers and paves the way for their widespread adoption.

physics.optics↗

arXiv2025

Parkes transient events: II. Pulsar single pulses database containing raw data segment

We have re-processed single pulse candidates from the first four years (1997-2001) of the Parkes Multibeam receiver system observations, creating a new Parkes transient database (PTD II) that contains 165,592 single pulses from 363 known pulsars. Unlike previous databases, PTD II preserves the critical raw data segments of each detected pulse, enabling detailed analyses of emission physics while maintaining a compact size of only 1.5 GB. The database employs a sqlite3 structure organising pulsar metadata, observation files, and pulse events with their raw data stored in binary format. We provide processing tools for extracting and analysing single-pulse data, enabling fluence fitting and statistical analysis. Our pulsars exhibit diverse fluence distributions, such as log-normal, Gaussian, and unimodal. Temporal analyses reveal significant evolution in emission characteristics of several pulsars, including event rate variations spanning two orders of magnitude in PSR J1602-5100 and PSR J0942-5552. Beyond supporting fundamental pulsar emission studies, the database serves as a valuable training resource for developing new single-pulse detection algorithms in the presence of radio frequency interference.

astro-ph.HE↗

arXiv2025

SafeAuto: Knowledge-Enhanced Safe Autonomous Driving with Multimodal Foundation Models

Traditional autonomous driving systems often struggle to connect high-level reasoning with low-level control, leading to suboptimal and sometimes unsafe behaviors. Recent advances in multimodal large language models (MLLMs), which process both visual and textual data, offer an opportunity to unify perception and reasoning. However, effectively embedding precise safety knowledge into MLLMs for autonomous driving remains a significant challenge. To address this, we propose SafeAuto, a framework that enhances MLLM-based autonomous driving by incorporating both unstructured and structured knowledge. First, we introduce a Position-Dependent Cross-Entropy (PDCE) loss to improve low-level control signal predictions when values are represented as text. Second, to explicitly integrate safety knowledge, we develop a reasoning component that translates traffic rules into first-order logic (e.g., "red light $\implies$ stop") and embeds them into a probabilistic graphical model (e.g., Markov Logic Network) to verify predicted actions using recognized environmental attributes. Additionally, our Multimodal Retrieval-Augmented Generation (RAG) model leverages video, control signals, and environmental attributes to learn from past driving experiences. Integrating PDCE, MLN, and Multimodal RAG, SafeAuto outperforms existing baselines across multiple datasets, enabling more accurate, reliable, and safer autonomous driving. The code is available at https://github.com/AI-secure/SafeAuto.

cs.RO↗

arXiv2025

Bright bursts with sub-millisecond structures of FRB 20230607A in a highly magnetized environment

We report the observations of a repeating FRB 20230607A for 15.6 hours spanning 16 months using the Five-hundred-meter Aperture Spherical Radio Telescope (FAST) with the detection of 565 bursts. We present three bright bursts with detailed temporal/spectral structures. We also report that one burst carries a narrow component with a width of only 0.3 ms, which is surrounded by broader components. This suggests that repeaters can make both narrow and broad components in one burst. With the narrow spike, we precisely measure the dispersion measure (DM) of $362.85 \pm 0.15 \;{\rm pc\,cm^{-3}}$ and the Faraday rotation measures (RMs) of and $-12249.0\pm 1.5 \; {\rm rad\,m^{-2}}$. We also analyze the statistical distribution of the burst parameters, including waiting times, temporal widths, central frequencies and frequency widths, fluences and energies, all showing typical distributions of known active repeaters. In particular, most bursts show narrow spectra with $Δν/ν_0 = 0.125\pm 0.001$. This fact, together with the narrow 0.3 ms spike, strongly suggests a magnetospheric origin of the FRB emission. Based on a predicted correlation between RM and the luminosity of a persistent radio source (PRS) by Yang et al., we predict that PRS should have a specific luminosity of the order of $10^{29} \ {\rm erg \ s^{-1} \ Hz^{-1}}$ and encourage a search for such a PRS.

astro-ph.HE↗

arXiv2025

Quasi-periodic sub-structure of RRAT J1913+1330

Recent findings suggest a universal relationship between the quasi-periodic sub-structures and rotational periods across various types of radio-emitting neutron stars. In this study, we report the detection of 12 quasi-periodic sub-structures in a rotating radio transient (RRAT) J1913+1330 using the Five-hundred-meter Aperture Spherical Radio Telescope (FAST). This is the second known RRAT exhibiting quasi-periodic sub-structures. Our result reinforces the observed relationship between quasi-periodicity and rotational period. The polarization analysis reveals that 11 of the 12 pulses exhibit high linear polarization consistent with the quasi-periodic behaviour of the total intensity, while circular polarization with detectable quasi-periodic sub-structures is observed in only three pulses. No correlation is found between the sub-structure periods and their widths, peak fluxes, or fluences, even under the extremely variable single-pulse energy and morphology observed in J1913+1330.

astro-ph.HE↗

arXiv2025

Unified Autoregressive Visual Generation and Understanding with Continuous Tokens

We present UniFluid, a unified autoregressive framework for joint visual generation and understanding leveraging continuous visual tokens. Our unified autoregressive architecture processes multimodal image and text inputs, generating discrete tokens for text and continuous tokens for image. We find though there is an inherent trade-off between the image generation and understanding task, a carefully tuned training recipe enables them to improve each other. By selecting an appropriate loss balance weight, the unified model achieves results comparable to or exceeding those of single-task baselines on both tasks. Furthermore, we demonstrate that employing stronger pre-trained LLMs and random-order generation during training is important to achieve high-fidelity image generation within this unified framework. Built upon the Gemma model series, UniFluid exhibits competitive performance across both image generation and understanding, demonstrating strong transferability to various downstream tasks, including image editing for generation, as well as visual captioning and question answering for understanding.

cs.CV↗

arXiv2025

Step-Video-T2V Technical Report: The Practice, Challenges, and Future of Video Foundation Model

We present Step-Video-T2V, a state-of-the-art text-to-video pre-trained model with 30B parameters and the ability to generate videos up to 204 frames in length. A deep compression Variational Autoencoder, Video-VAE, is designed for video generation tasks, achieving 16x16 spatial and 8x temporal compression ratios, while maintaining exceptional video reconstruction quality. User prompts are encoded using two bilingual text encoders to handle both English and Chinese. A DiT with 3D full attention is trained using Flow Matching and is employed to denoise input noise into latent frames. A video-based DPO approach, Video-DPO, is applied to reduce artifacts and improve the visual quality of the generated videos. We also detail our training strategies and share key observations and insights. Step-Video-T2V's performance is evaluated on a novel video generation benchmark, Step-Video-T2V-Eval, demonstrating its state-of-the-art text-to-video quality when compared with both open-source and commercial engines. Additionally, we discuss the limitations of current diffusion-based model paradigm and outline future directions for video foundation models. We make both Step-Video-T2V and Step-Video-T2V-Eval available at https://github.com/stepfun-ai/Step-Video-T2V. The online version can be accessed from https://yuewen.cn/videos as well. Our goal is to accelerate the innovation of video foundation models and empower video content creators.

cs.CV↗

arXiv2025

Inference-Time Scaling for Diffusion Models beyond Scaling Denoising Steps

Generative models have made significant impacts across various domains, largely due to their ability to scale during training by increasing data, computational resources, and model size, a phenomenon characterized by the scaling laws. Recent research has begun to explore inference-time scaling behavior in Large Language Models (LLMs), revealing how performance can further improve with additional computation during inference. Unlike LLMs, diffusion models inherently possess the flexibility to adjust inference-time computation via the number of denoising steps, although the performance gains typically flatten after a few dozen. In this work, we explore the inference-time scaling behavior of diffusion models beyond increasing denoising steps and investigate how the generation performance can further improve with increased computation. Specifically, we consider a search problem aimed at identifying better noises for the diffusion sampling process. We structure the design space along two axes: the verifiers used to provide feedback, and the algorithms used to find better noise candidates. Through extensive experiments on class-conditioned and text-conditioned image generation benchmarks, our findings reveal that increasing inference-time compute leads to substantial improvements in the quality of samples generated by diffusion models, and with the complicated nature of images, combinations of the components in the framework can be specifically chosen to conform with different application scenario.

cs.CV↗

arXiv2025

Full C- and L-band tunable erbium-doped integrated lasers via scalable manufacturing

Erbium (Er) ions are the gain medium of choice for fiber-based amplifiers and lasers, offering a long excited-state lifetime, slow gain relaxation, low amplification nonlinearity and noise, and temperature stability compared to semiconductor-based platforms. Recent advances in ultra-low-loss silicon nitride (Si$_3$N$_4$) photonic integrated circuits, combined with ion implantation, have enabled the realization of high-power on-chip Er amplifiers and lasers with performance comparable to fiber-based counterparts, supporting compact photonic systems. Yet, these results are limited by the high (2 MeV) implantation beam energy required for tightly confined Si$_3$N$_4$ waveguides (700 nm height), preventing volume manufacturing of Er-doped photonic integrated circuits. Here, we overcome these limitations and demonstrate the first fully wafer-scale, foundry-compatible Er-doped photonic integrated circuit-based tunable lasers. Using 200 nm-thick Si$_3$N$_4$ waveguides, we reduce the ion beam energy requirement to below 500 keV, enabling efficient wafer-scale implantation with an industrial 300 mm ion implanter. We demonstrate a laser wavelength tuning range of 91 nm, covering nearly the entire optical C- and L-bands, with fiber-coupled output power reaching 36 mW and an intrinsic linewidth of 95 Hz. The temperature-insensitive properties of erbium ions allowed stable laser operation up to 125$^{\circ}$C and lasing with less than 15 MHz drift for over 6 hours at room temperature using a remote fiber pump. The fully scalable, low-cost fabrication of Er-doped waveguide lasers opens the door for widespread adoption in coherent communications, LiDAR, microwave photonics, optical frequency synthesis, and free-space communications.

physics.optics↗

arXiv2024

STNMamba: Mamba-based Spatial-Temporal Normality Learning for Video Anomaly Detection

Video anomaly detection (VAD) has been extensively researched due to its potential for intelligent video systems. However, most existing methods based on CNNs and transformers still suffer from substantial computational burdens and have room for improvement in learning spatial-temporal normality. Recently, Mamba has shown great potential for modeling long-range dependencies with linear complexity, providing an effective solution to the above dilemma. To this end, we propose a lightweight and effective Mamba-based network named STNMamba, which incorporates carefully designed Mamba modules to enhance the learning of spatial-temporal normality. Firstly, we develop a dual-encoder architecture, where the spatial encoder equipped with Multi-Scale Vision Space State Blocks (MS-VSSB) extracts multi-scale appearance features, and the temporal encoder employs Channel-Aware Vision Space State Blocks (CA-VSSB) to capture significant motion patterns. Secondly, a Spatial-Temporal Interaction Module (STIM) is introduced to integrate spatial and temporal information across multiple levels, enabling effective modeling of intrinsic spatial-temporal consistency. Within this module, the Spatial-Temporal Fusion Block (STFB) is proposed to fuse the spatial and temporal features into a unified feature space, and the memory bank is utilized to store spatial-temporal prototypes of normal patterns, restricting the model's ability to represent anomalies. Extensive experiments on three benchmark datasets demonstrate that our STNMamba achieves competitive performance with fewer parameters and lower computational costs than existing methods.

cs.CV↗

arXiv2024

SANPO: A Scene Understanding, Accessibility and Human Navigation Dataset

Vision is essential for human navigation. The World Health Organization (WHO) estimates that 43.3 million people were blind in 2020, and this number is projected to reach 61 million by 2050. Modern scene understanding models could empower these people by assisting them with navigation, obstacle avoidance and visual recognition capabilities. The research community needs high quality datasets for both training and evaluation to build these systems. While datasets for autonomous vehicles are abundant, there is a critical gap in datasets tailored for outdoor human navigation. This gap poses a major obstacle to the development of computer vision based Assistive Technologies. To overcome this obstacle, we present SANPO, a large-scale egocentric video dataset designed for dense prediction in outdoor human navigation environments. SANPO contains 701 stereo videos of 30+ seconds captured in diverse real-world outdoor environments across four geographic locations in the USA. Every frame has a high resolution depth map and 112K frames were annotated with temporally consistent dense video panoptic segmentation labels. The dataset also includes 1961 high-quality synthetic videos with pixel accurate depth and panoptic segmentation annotations to balance the noisy real world annotations with the high precision synthetic annotations. SANPO is already publicly available and is being used by mobile applications like Project Guideline to train mobile models that help low-vision users go running outdoors independently. To preserve anonymization during peer review, we will provide a link to our dataset upon acceptance. SANPO is available here: https://google-research-datasets.github.io/sanpo_dataset/

cs.CV↗

arXiv2024

LLM-RG4: Flexible and Factual Radiology Report Generation across Diverse Input Contexts

Drafting radiology reports is a complex task requiring flexibility, where radiologists tail content to available information and particular clinical demands. However, most current radiology report generation (RRG) models are constrained to a fixed task paradigm, such as predicting the full ``finding'' section from a single image, inherently involving a mismatch between inputs and outputs. The trained models lack the flexibility for diverse inputs and could generate harmful, input-agnostic hallucinations. To bridge the gap between current RRG models and the clinical demands in practice, we first develop a data generation pipeline to create a new MIMIC-RG4 dataset, which considers four common radiology report drafting scenarios and has perfectly corresponded input and output. Secondly, we propose a novel large language model (LLM) based RRG framework, namely LLM-RG4, which utilizes LLM's flexible instruction-following capabilities and extensive general knowledge. We further develop an adaptive token fusion module that offers flexibility to handle diverse scenarios with different input combinations, while minimizing the additional computational burden associated with increased input volumes. Besides, we propose a token-level loss weighting strategy to direct the model's attention towards positive and uncertain descriptions. Experimental results demonstrate that LLM-RG4 achieves state-of-the-art performance in both clinical efficiency and natural language generation on the MIMIC-RG4 and MIMIC-CXR datasets. We quantitatively demonstrate that our model has minimal input-agnostic hallucinations, whereas current open-source models commonly suffer from this problem.

cs.CL↗

arXiv2024

VideoGLUE: Video General Understanding Evaluation of Foundation Models

We evaluate the video understanding capabilities of existing foundation models (FMs) using a carefully designed experiment protocol consisting of three hallmark tasks (action recognition,temporal localization, and spatiotemporal localization), eight datasets well received by the community, and four adaptation methods tailoring an FM for downstream tasks. Furthermore, we jointly profile FMs' efficacy and efficiency when adapting to general video understanding tasks using cost measurements during both training and inference. Our main findings areas follows. First, task-specialized models significantly outperform the seven FMs studied in this work, in sharp contrast to what FMs have achieved in natural language and image understanding. Second, video-native FMs, whose pretraining data mainly contains the video modality, are generally better than image-native FMs in classifying motion-rich videos, localizing actions in time, and understanding a video of more than one action. Third, the video-native FMs can perform well on video tasks under light adaptations to downstream tasks (e.g., freezing the FM backbones), while image-native FMs win in full end-to-end finetuning. The first two observations reveal the need and tremendous opportunities to conduct research on video-focused FMs, and the last confirms that both tasks and adaptation methods matter when it comes to the evaluation of FMs. Our code is released under: https://github.com/tensorflow/models/tree/master/official/projects/videoglue.

cs.CV↗

arXiv2024

Short-Term Photovoltaic Forecasting Model for Qualifying Uncertainty during Hazy Weather

Solar energy is one of the most promising renewable energy resources. Forecasting photovoltaic power generation is an important way to increase photovoltaic penetration. However, the difficulty in qualifying the uncertainty of PV power generation, especially during hazy weather, makes forecasting challenging. This paper proposes a novel model to address the issue. We introduce a modified entropy to qualify uncertainty during hazy weather while clustering and attention mechanisms are employed to reduce computational costs and enhance forecasting accuracy, respectively. Hyperparameters were adjusted using an optimization algorithm. Experiments on two datasets related to hazy weather demonstrate that our model significantly improves forecasting accuracy compared to existing models.

cs.LG↗

arXiv2024

The Giant Radio Array for Neutrino Detection (GRAND) Collaboration -- Contributions to the 38th International Cosmic Ray Conference (ICRC 2023)

The Giant Radio Array for Neutrino Detection (GRAND) is an envisioned observatory of ultra-high-energy particles of cosmic origin, with energies in excess of 100 PeV. GRAND uses large surface arrays of autonomous radio-detection units to look for the radio emission from extensive air showers that are triggered by the interaction of ultra-high-energy cosmic rays, gamma rays, and neutrinos in the atmosphere or underground. In particular, for ultra-high-energy neutrinos, the future final phase of GRAND aims to be sensitive enough to discover them in spite of their plausibly tiny flux. Presently, three prototype GRAND radio arrays are in operation: GRANDProto300, in China, GRAND@Auger, in Argentina, and GRAND@Nancay, in France. Their goals are to field-test the design of the radio-detection units, understand the radio background to which they are exposed, and develop tools for diagnostic, data gathering, and data analysis. This list of contributions to the 38th International Cosmic Ray Conference (ICRC 2023) presents an overview of GRAND, in its present and future incarnations, and a look at the first data collected by GRANDProto13, the first phase of GRANDProto300.

hep-ex↗

arXiv2024

VideoPoet: A Large Language Model for Zero-Shot Video Generation

We present VideoPoet, a language model capable of synthesizing high-quality video, with matching audio, from a large variety of conditioning signals. VideoPoet employs a decoder-only transformer architecture that processes multimodal inputs -- including images, videos, text, and audio. The training protocol follows that of Large Language Models (LLMs), consisting of two stages: pretraining and task-specific adaptation. During pretraining, VideoPoet incorporates a mixture of multimodal generative objectives within an autoregressive Transformer framework. The pretrained LLM serves as a foundation that can be adapted for a range of video generation tasks. We present empirical results demonstrating the model's state-of-the-art capabilities in zero-shot video generation, specifically highlighting VideoPoet's ability to generate high-fidelity motions. Project page: http://sites.research.google/videopoet/

cs.CV↗

arXiv2024

The Performance of AC-coupled Strip LGAD developed by IHEP

The AC-coupled Strip LGAD (Strip AC-LGAD) is a novel LGAD design that diminishes the density of readout electronics through the use of strip electrodes, enabling the simultaneous measurement of time and spatial information. The Institute of High Energy Physics has designed a long Strip AC-LGAD prototype with a strip electrode length of 5.7 mm and pitches of 150 $μm$, 200 $μm$, and 250 $μm$. Spatial and timing resolutions of the long Strip AC-LGAD are studied by pico-second laser test and beta source tests. The laser test demonstrates that spatial resolution improves as the pitch size decreases, with an optimal resolution achieved at 8.3 $μ$m. Furthermore, the Beta source test yields a timing resolution of 37.6 ps.

physics.ins-det↗

arXiv2024

Development of the strip LGAD detector with double-end readout for future colliders

The Low-Gain Avalanche Diode (LGAD) is a new silicon detector and holds wide application prospects in particle physics experiments due to its excellent timing resolution. The LGAD with a pixel size of 1.3 mm $\times$ 1.3 mm was used to construct a High Granularity Timing Detector (HGTD) in ATLAS experiments to solve the pile-up problem. Meanwhile, the Circular Electron Positron Collider (CEPC) also proposes detectors using the LGAD. However, pixel LGAD exhibits higher readout electronics density and cost, which somewhat limits the application of LGADs. To decrease the readout electronics density, the Institute of High Energy Physics (IHEP) of the Chinese Academy of Sciences has designed strip LGADs with larger areas. These strip LGADs are all 19 mm in length but with different widths of 1.0 mm, 0.5 mm, and 0.3 mm. This article provides a detailed introduction to the design parameters of these strip LGADs and tests their electrical characteristics, including leakage current, break-down voltage, depletion capacitance, etc. The timing resolution and signal-to-noise ratio of the three strip LGAD sensors were investigated using a beta source test system. The position resolution parallel to the strip direction was tested and analyzed for the first time using a pico-second laser test system. Tests have demonstrated that the timing resolution of strip LGADs can reach about 37.5 ps, and position resolution parallel to the strip direction is better than 1 mm.

physics.ins-det↗