Searcharxiv⌕ Search

arXiv subjects

Qiangang Wang

Publications and source records attributed to Qiangang Wang.

3 recordsLinked to original sources

Early Spectroscopy of the Type II SN 2024phv Suggests Short-Lived Circumstellar Interaction

We present the high-cadence light curve and rapid spectroscopy of the Type II SN 2024phv, discovered just 16.8 hours after the estimated explosion time. The early light curve exhibits a rapid 2 mag rise over the first day of observation, and spectra at 1.7 and 2.6 days reveal weak, narrow emission lines, suggesting moderate interaction with circumstellar material during the first few days after explosion. Models of shock-cooling emission provide a reasonable fit to the light curve after day 2, implying minimal interaction at later epochs, and yield a best-fit progenitor radius of $R=390\pm90\,R_{\odot}$. We also compare our earliest spectra with radiation hydrodynamical models of exploding red supergiants to measure circumstellar densities of ${\sim}10^{-12}$ g cm$^{-3}$ within a few stellar radii of the progenitor, followed by a sharp dropoff in the density profile. This rapidly decreasing profile leads us to conclude that the progenitor of SN 2024phv had an extended atmosphere and/or closely confined circumstellar material, likely from rapidly increasing or eruptive mass loss prior to explosion. In the context of other supernovae discovered at similarly early phases, SN 2024phv points to a diversity in the mass-loss rates of red supergiant stars in their final years.

astro-ph.HE↗

Speed by Simplicity: A Single-Stream Architecture for Fast Audio-Video Generative Foundation Model

We present daVinci-MagiHuman, an open-source audio-video generative foundation model for human-centric generation. daVinci-MagiHuman jointly generates synchronized video and audio using a single-stream Transformer that processes text, video, and audio within a unified token sequence via self-attention only. This single-stream design avoids the complexity of multi-stream or cross-attention architectures while remaining easy to optimize with standard training and inference infrastructure. The model is particularly strong in human-centric scenarios, producing expressive facial performance, natural speech-expression coordination, realistic body motion, and precise audio-video synchronization. It supports multilingual spoken generation across Chinese (Mandarin and Cantonese), English, Japanese, Korean, German, and French. For efficient inference, we combine the single-stream backbone with model distillation, latent-space super-resolution, and a Turbo VAE decoder, enabling generation of a 5-second 256p video in 2 seconds on a single H100 GPU. In automatic evaluation, daVinci-MagiHuman achieves the highest visual quality and text alignment among leading open models, along with the lowest word error rate (14.60%) for speech intelligibility. In pairwise human evaluation, it achieves win rates of 80.0% against Ovi 1.1 and 60.9% against LTX 2.3 over 2000 comparisons. We open-source the complete model stack, including the base model, the distilled model, the super-resolution model, and the inference codebase.

cs.CV↗

Long-Context Attention Benchmark: From Kernel Efficiency to Distributed Context Parallelism

Transformer-based large language models (LLMs) have achieved remarkable success, yet their standard attention mechanism incurs quadratic computation and memory costs with respect to sequence length, posing a major bottleneck for long-context training. Prior work tackles this challenge along two directions: (1) kernel-level optimizations, which accelerate dense and sparse attention operators; and (2) module-level strategies, often referred to as distributed attention or context parallel training, which scale attention across multiple devices. However, systematic evaluation still remains limited: operator-level comparisons are often incomplete, while context parallel strategies are typically framework-specific, with unclear performance analysis across contexts. To address these gaps, we propose a unified benchmark that integrates representative attention kernels and context parallel mechanisms with a modular and extensible interface for evaluation. The benchmark evaluates methods along two critical dimensions: (1) attention mask patterns, which strongly affect efficiency, scalability, and usability, and (2) sequence length and distributed scale, which determine performance under extreme long-context training. Through comprehensive experiments on the cluster of up to 96 GPUs, our benchmark enables reproducible comparisons, highlights method-specific trade-offs, and provides practical guidance for designing and deploying attention mechanisms in long-context LLM training.

cs.LG↗