arXiv subjects

Lang Zhao

Publications and source records attributed to Lang Zhao.

4 recordsLinked to original sources

arXiv2026

Near-extremal critical dynamics driven by a phantom field in Einstein-Maxwell-scalar theory

We investigate the fully nonlinear dynamics of Einstein-Maxwell-scalar black holes in the low-mass, near-extremal region of the solution space, which is difficult to access with conventional positive-energy perturbations. An independent phantom field is introduced as a controllable negative-energy channel, and its initial amplitude is varied while the remaining parameters of the initial data are held fixed. The phantom contribution is concentrated during the early-time infall and reduces the quasi-local mass in the black-hole region at fixed electric charge, thereby allowing the evolution to approach near-extremal configurations. Near a critical amplitude $p_\ast$, evolutions on both sides of the threshold enter the same long-lived scalarized intermediate stage before separating toward either a near-extremal Reissner-Nordstrom black hole or a nonextremal hairy black hole. The residence time near this intermediate stage obeys the logarithmic scaling law of Type I critical behavior, and the growth rate extracted from the departure is consistent with the inverse lifetime coefficient. The intermediate stage is characterized by a small horizon surface gravity and finite horizon curvature, with no curvature blow-up resolved at the horizon during the simulated interval. These results demonstrate that phantom-assisted negative-energy transport provides a dynamical route into the near-extremal sector of the EMS solution space and reveals a threshold-mediated transition between bald and hairy black-hole end states.

gr-qc↗

arXiv2026

FSVL: Taxonomic Fast-Slow Vision-to-Language Collaborative Framework for Interpretable Confusable Fish Species Identification and Knowledge Generation in the Middle Yangtze River

The Yangtze River Basin is a critical region for freshwater biodiversity conservation. The implementation of the Yangtze River Ten-Year Fishing Ban has increased the importance of effective aquatic biodiversity monitoring and species-level identification. However, fine-grained identification of closely related freshwater fish species remains challenging due to subtle morphological differences. Existing deep learning methods often struggle to balance classification accuracy and inference efficiency, while LLMs may generate unsupported information when producing specialized biological knowledge. To address these challenges, this study proposes FSVL, a fast-slow dual-pathway framework integrating visual recognition with LLM-based Retrieval-Augmented Generation (RAG). A lightweight ResNet50 fast path performs preliminary classification across 51 fish species in the Middle Yangtze River Basin and routes predefined groups of easily confused species to a slow path for refined identification. Guided by taxonomic criteria, the slow path uses YOLOv8n to detect key morphological regions, including the head, dorsal fin, and tail, leveraging morphometric and outline traits to improve discrimination among closely related species. The framework further integrates a retrieval-enhanced Qwen3-14B backend, using soft tokens and Low-Rank Adaptation (LoRA) for parameter-efficient domain knowledge generation. On the evaluated test set of 607 images, the fast classifier achieved 99.67% accuracy, whereas the cascade corrected the two errors, yielding 100.00% overall accuracy; further validation on independent specimens and external conditions remains needed. Across 255 knowledge generation samples, scientific name accuracy rises from 91.37% to 100.00%, genus accuracy from 90.20% to 98.82%, and ROUGE-L from 0.335 to 0.655, supporting intelligent biodiversity monitoring and fisheries management.

cs.CV↗

arXiv2025

Radial Oscillations of Scalar Hair in Black Hole Bombs

Recent research has revealed a novel nonlinear mechanism, distinct from the linear superradiant instability, which triggers the black hole bomb phenomenon. Introducing a massive complex scalar field with nonlinear self-interactions drives the Reissner-Nordström black hole to shed substantial energy, thereby triggering a black hole bomb. Radial oscillations in the scalar hair profile are observed during this process. In this paper, we further reveal that physical quantities associated with scalar hair exhibit identical oscillation patterns during the evolution of the black hole-scalar field system. Moreover, the oscillation frequency exhibits a linear dependence on the gauge coupling constant of the scalar field with other parameters fixed. Meanwhile, the horizon radius of hairy black holes and the mass within the horizon increase monotonically with the gauge coupling constant. We have also identified a critical initial charge value that distinguishes hairy solutions that trigger black hole bombs from those that do not.

gr-qc↗

arXiv2020

Spatial Damage Characterization in Self-Sensing Materials via Neural Network-Aided Electrical Impedance Tomography: A Computational Study

Continuous structural health monitoring (SHM) and integrated nondestructive evaluation (NDE) are important for ensuring the safe operation of high-risk engineering structures. Recently, piezoresistive nanocomposite materials have received much attention for SHM and NDE. These materials are self-sensing because their electrical conductivity changes in response to deformation and damage. Combined with electrical impedance tomography (EIT), it is possible to map deleterious effects. However, EIT suffers from important limitations -- it is computationally expensive, provides indistinct information on damage shape, and can miss multiple damages if they are close together. In this article we apply a novel neural network approach to quantify damage metrics such as size, number, and location from EIT data. This network is trained using a simulation routine calibrated to experimental data for a piezoresistive carbon nanofiber-modified epoxy. Our results show that the network can predict the number of damages with 99.2% accuracy, quantify damage size with respect to the averaged radius at an average of 2.46% error, and quantify damage position with respect to the domain length at an average of 0.89% error. These results are an important first step in translating the combination of self-sensing materials and EIT to real-world SHM and NDE.

eess.IV↗