arXiv subjects

Ruihan Zhang

Publications and source records attributed to Ruihan Zhang.

At least 19 recordsLinked to original sources

arXiv2026

Facility integration of the NASA IRTF adaptive secondary mirror

IRTF-ASM-1 has been functioning well since its first light in 2024. This adaptive secondary mirror (ASM) was primarily developed to be an on-sky demonstration of the new hybrid variable reluctance actuator technology at the NASA Infrared Telescope Facility (IRTF). However, due to its physical robustness and our previous demonstrations of sensitivity enhancements with the ASM, we are interested in using it to optimize telescope image quality on a nightly basis. This will directly benefit science observations. However, as IRTF does not currently have adaptive optics expertise, we have been developing the system to be used with minimal human intervention. We present our progress in developing software for active optics mode with IRTF-ASM-1 using the single conjugate, facility 2x2 Shack-Hartmann wavefront sensor Felix. We also present techniques for removing low-order, large amplitude non-common path aberrations between Felix and our science instruments.

astro-ph.IM↗

arXiv2026

Progress report on the integration of the IRTF adaptive secondary mirror

IRTF-ASM-1 is the first on-sky adaptive secondary mirror using the hybrid variable reluctance (HVR) actuators developed by the Netherlands Organization for Applied Scientific Research (TNO). Since its first light in April 2024, the ASM has continued to work consistently well with no hardware issues. The primary purpose of IRTF-ASM-1 is to serve as a demonstration of the HVR actuator technology, both in terms of verifying its robustness and testing calibration methods that are relevant to larger ASMs. However, as the ASM has proven to be easy to handle and reliable in its performance, we are moving toward integrating the ASM for long-term use at IRTF. We present closed loop results with the ASM and IRTF's off-axis facility wavefront sensor FELIX. Correcting the first seven Zernike modes through coma at 90-180 Hz, we are able to enhance the seeing by a factor of 1.8 in FWHM under 0."5 seeing conditions. We also performed the first science observations with the ASM in "static" mode and demonstrated that we can improve the throughput of IRTF's slit spectrograph by approximately a factor of 2, although this requires good reference slopes in FELIX. In the near future, we plan to optimize the calibration of reference slopes in FELIX and streamline the software so that the system can be operated by a non-AO expert.

astro-ph.IM↗

arXiv2026

Robo-AO-2: entering the era of automated science operations, hybrid wavefront sensing, and adaptive secondary integration

We present the first science results and new technical milestones from the Robo-AO-2 facility at the University of Hawaii 2.2-m telescope. Following successful commissioning, the system began science operations in 2025. We are starting a large-scale survey of young stars in the Scorpius-Centaurus association to detect sub-stellar companions, vetting the Habitable Worlds Observatory Target Stars and Systems list, and discovering stellar blends for TESS exoplanet host candidates. We report on the commissioning of the natural guide star wavefront sensor, supporting science and future hybrid laser-stellar sensing demonstrations. Finally, we detail the automation of the telescope facility, Robo88, via updated telescope control systems and absolute encoders, and the integration of Robo-AO-2 with the telescope's new adaptive secondary mirror.

astro-ph.IM↗

arXiv2026

The $\mu$ Herculis system solved after nearly three centuries

$\mu$ Herculis is a bright, nearby quadruple system. Its brightest member, $\mu$ Her Aa, displays solar-like oscillations, establishing the system as a crucial benchmark for asteroseismology, provided that its mass can be determined independently of stellar models. We aim to resolve the full hierarchical architecture of the system and determine precise, model-independent dynamical masses for all four components (Aa, Ab, B, and C), along with a consistent astrometric solution for the system's centre of mass. We performed a joint fit of radial velocities, relative astrometry and absolute astrometry from \textit{Hipparcos}, \textit{Gaia} DR3, and ground-based catalogues, spanning nearly three centuries. Our forward-modelling framework simultaneously constrains the Keplerian orbits of the inner Aa--Ab and B--C subsystems, the wide A--BC orbit, and the sky motion and parallax of the total centre of mass. Leveraging several complementary datasets and the decisive 2023 periastron passage of the Aa--Ab pair, we precisely determine all orbital parameters and obtain sub-percent precision on the component masses: $M_{\rm Aa} = 1.134 \pm 0.007\,M_{\odot}$, $M_{\rm Ab} = 0.2286 \pm 0.0006\,M_{\odot}$, $M_{\rm C} = 0.445 \pm 0.005\,M_{\odot}$, and $M_{\rm B} = 0.417 \pm 0.005\,M_{\odot}$. We derive a system parallax of $\varpi_{\rm CM} = 120.069 \pm 0.089\,\mathrm{mas}$ that reconciles and improves upon the individual \textit{Hipparcos} and \textit{Gaia} DR3 values.

astro-ph.SR↗

arXiv2026

Majorana Zero Modes and Topological Nature in Bi2Ta3S6-family Superconductors

In this work, we report that Bi2Ta3S6-family superconductors exhibit nontrivial band topology. They possess a natural quantum-well structure consisting of alternating stacks of TaS2 and honeycomb Bi layers, which contribute superconducting and topological properties, respectively. Symmetry-based indicators $(\mathbb{Z}_4;\mathbb{Z}_{2}\mathbb{Z}_{2}\mathbb{Z}_{2})=(2;000)$ reveal that the topological nature arises entirely from the Bi layers, which belong to a quantum spin Hall phase characterized by a $p_x-p_y$ model on a honeycomb lattice. The topological zigzag (ZZ) and armchair (AC) edge states are obtained. Using VASP2KP, the in-plane $g$ factors of these topological edge states are computed from the ab initio calculations: $g_{x/y}^{\mathrm{ZZ}}=2.07/1.60$ and $g_{x/y}^{\mathrm{AC}}=0.50/0.06$. The strong anisotropy of the edge-state $g$ factors allows us to explore Majorana zero modes in the Bi monolayer on a superconductor, which can be obtained by exfoliation or molecular beam epitaxy. The relaxed structures of the Bi2Ta3Se6, Bi2Nb3S6 and Bi2Nb3Se6 are obtained. Their superconducting transition temperature $T_c$ are estimated based on the electron-phonon coupling and the McMillan formula. Furthermore, using the experimental superconducting gap $\Delta$ and the computed $g$ factors, we obtain the phase diagram, which shows that the in-plane field $B_y>2.62\mathrm{ T}$ can generate corner Majorana zero modes in the Bi monolayer of the superconductor Bi2Ta3S6. A similar paradigm also applies to the Bi2Ta3S6 bulk with the emergence of Majorana hinge states. These natural quantum-well superconductors therefore offer ideal platforms for exploring topological superconductivity and Majorana zero modes.

cond-mat.supr-con↗

arXiv2026

Rendering Data Unlearnable by Exploiting LLM Alignment Mechanisms

Large language models (LLMs) are increasingly trained on massive, heterogeneous text corpora, raising serious concerns about the unauthorised use of proprietary or personal data during model training. In this work, we address the problem of data protection against unwanted model learning in a realistic black-box setting. We propose Disclaimer Injection, a novel data-level defence that renders text unlearnable to LLMs. Rather than relying on model-side controls or explicit data removal, our approach exploits the models' own alignment mechanisms: by injecting carefully designed alignment-triggering disclaimers to prevent effective learning. Through layer-wise analysis, we find that fine-tuning on such protected data induces persistent activation of alignment-related layers, causing alignment constraints to override task learning even on common inputs. Consequently, models trained on such data exhibit substantial and systematic performance degradation compared to standard fine-tuning. Our results identify alignment behaviour as a previously unexplored lever for data protection and, to our knowledge, present the first practical method for restricting data learnability at LLM scale without requiring access to or modification of the training pipeline.

cs.CL↗

arXiv2025

Towards Provably Unlearnable Examples via Bayes Error Optimisation

The recent success of machine learning models, especially large-scale classifiers and language models, relies heavily on training with massive data. These data are often collected from online sources. This raises serious concerns about the protection of user data, as individuals may not have given consent for their data to be used in training. To address this concern, recent studies introduce the concept of unlearnable examples, i.e., data instances that appear natural but are intentionally altered to prevent models from effectively learning from them. While existing methods demonstrate empirical effectiveness, they typically rely on heuristic trials and lack formal guarantees. Besides, when unlearnable examples are mixed with clean data, as is often the case in practice, their unlearnability disappears. In this work, we propose a novel approach to constructing unlearnable examples by systematically maximising the Bayes error, a measurement of irreducible classification error. We develop an optimisation-based approach and provide an efficient solution using projected gradient ascent. Our method provably increases the Bayes error and remains effective when the unlearning examples are mixed with clean samples. Experimental results across multiple datasets and model architectures are consistent with our theoretical analysis and show that our approach can restrict data learnability, effectively in practice.

cs.AI↗

arXiv2025

Visual Evaluative AI: A Hypothesis-Driven Tool with Concept-Based Explanations and Weight of Evidence

This paper presents Visual Evaluative AI, a decision aid that provides positive and negative evidence from image data for a given hypothesis. This tool finds high-level human concepts in an image and generates the Weight of Evidence (WoE) for each hypothesis in the decision-making process. We apply and evaluate this tool in the skin cancer domain by building a web-based application that allows users to upload a dermatoscopic image, select a hypothesis and analyse their decisions by evaluating the provided evidence. Further, we demonstrate the effectiveness of Visual Evaluative AI on different concept-based explanation approaches.

cs.CV↗

arXiv2025

Correct-By-Construction: Certified Individual Fairness through Neural Network Training

Fairness in machine learning is more important than ever as ethical concerns continue to grow. Individual fairness demands that individuals differing only in sensitive attributes receive the same outcomes. However, commonly used machine learning algorithms often fail to achieve such fairness. To improve individual fairness, various training methods have been developed, such as incorporating fairness constraints as optimisation objectives. While these methods have demonstrated empirical effectiveness, they lack formal guarantees of fairness. Existing approaches that aim to provide fairness guarantees primarily rely on verification techniques, which can sometimes fail to produce definitive results. Moreover, verification alone does not actively enhance individual fairness during training. To address this limitation, we propose a novel framework that formally guarantees individual fairness throughout training. Our approach consists of two parts, i.e., (1) provably fair initialisation that ensures the model starts in a fair state, and (2) a fairness-preserving training algorithm that maintains fairness as the model learns. A key element of our method is the use of randomised response mechanisms, which protect sensitive attributes while maintaining fairness guarantees. We formally prove that this mechanism sustains individual fairness throughout the training process. Experimental evaluations confirm that our approach is effective, i.e., producing models that are empirically fair and accurate. Furthermore, our approach is much more efficient than the alternative approach based on certified training (which requires neural network verification during training).

cs.LG↗

arXiv2025

A Test Suite for Efficient Robustness Evaluation of Face Recognition Systems

Face recognition is a widely used authentication technology in practice, where robustness is required. It is thus essential to have an efficient and easy-to-use method for evaluating the robustness of (possibly third-party) trained face recognition systems. Existing approaches to evaluating the robustness of face recognition systems are either based on empirical evaluation (e.g., measuring attacking success rate using state-of-the-art attacking methods) or formal analysis (e.g., measuring the Lipschitz constant). While the former demands significant user efforts and expertise, the latter is extremely time-consuming. In pursuit of a comprehensive, efficient, easy-to-use and scalable estimation of the robustness of face recognition systems, we take an old-school alternative approach and introduce RobFace, i.e., evaluation using an optimised test suite. It contains transferable adversarial face images that are designed to comprehensively evaluate a face recognition system's robustness along a variety of dimensions. RobFace is system-agnostic and still consistent with system-specific empirical evaluation or formal analysis. We support this claim through extensive experimental results with various perturbations on multiple face recognition systems. To our knowledge, RobFace is the first system-agnostic robustness estimation test suite.

cs.SE↗

arXiv2025

Generative AI for Film Creation: A Survey of Recent Advances

Generative AI (GenAI) is transforming filmmaking, equipping artists with tools like text-to-image and image-to-video diffusion, neural radiance fields, avatar generation, and 3D synthesis. This paper examines the adoption of these technologies in filmmaking, analyzing workflows from recent AI-driven films to understand how GenAI contributes to character creation, aesthetic styling, and narration. We explore key strategies for maintaining character consistency, achieving stylistic coherence, and ensuring motion continuity. Additionally, we highlight emerging trends such as the growing use of 3D generation and the integration of real footage with AI-generated elements. Beyond technical advancements, we examine how GenAI is enabling new artistic expressions, from generating hard-to-shoot footage to dreamlike diffusion-based morphing effects, abstract visuals, and unworldly objects. We also gather artists' feedback on challenges and desired improvements, including consistency, controllability, fine-grained editing, and motion refinement. Our study provides insights into the evolving intersection of AI and filmmaking, offering a roadmap for researchers and artists navigating this rapidly expanding field.

cs.CV↗

arXiv2024

Axion insulator, Weyl points, quantum anomalous Hall effect and magnetic topological phase transition in Eu3In2As4

The magnetic topological phases attract much interest, such as the axion insulator, higher-order topology, Weyl semimetals, and the quantum anomalous Hall effect (QAHE). Here, we predict that the axion insulator phase, magnetic Weyl points, and QAHE can be achieved in Eu3In2As4. Recently, the single-crystal Eu3In2As4 has been successfully synthesized, which exhibits an antiferromagnetic (AFM) ground state. Our first-principles calculations show that it lies on the phase boundary between multiple magnetic topological phases, and the magnetic anisotropy is weak, with an energy difference less than 1 meV. In the AFM state, it can be tuned to an axion insulator by tensile strain. The quantized axion angle $θ= π$ and the magnetic higher-order topology are characterized by the parity index $Z_4 = 2$. By applying an external magnetic field, the induced ferromagnetic (FM) state becomes an ideal magnetic topological semimetal with a single pair of Weyl points or a nodal ring. The QAHE can be achieved in FM multilayer films of Eu3In2As4 on a magnetic insulating substrate.

cond-mat.mtrl-sci↗

arXiv2024

Excitonic Instability in Ta2Pd3Te5 Monolayer

By systematic theoretical calculations, we have revealed an excitonic insulator (EI) in the Ta2Pd3Te5 monolayer. The bulk Ta2Pd3Te5 is a van der Waals (vdW) layered compound, whereas the vdW layer can be obtained through exfoliation or molecular-beam epitaxy. First-principles calculations show that the monolayer is a nearly zero-gap semiconductor with the modified Becke-Johnson functional. Due to the same symmetry of the band-edge states, the two-dimensional polarization $α_{2D}$ would be finite as the band gap goes to zero, allowing for an EI state in the compound. Using the first-principles many-body perturbation theory, the GW plus Bethe-Salpeter equation calculation reveals that the exciton binding energy is larger than the single-particle band gap, indicating the excitonic instability. The computed phonon spectrum suggests that the monolayer is dynamically stable without lattice distortion. Our findings suggest that the Ta2Pd3Te5 monolayer is an excitonic insulator without structural distortion.

cond-mat.mtrl-sci↗

arXiv2024

Simulation results for Robo-AO-2 using HAPA: a wavefront sensing technique for improving the adaptive optics correction of fainter stars

Direct imaging of exoplanets allows us to measure positions and chemical signatures of exoplanets. Given the limited resources for space observations where the atmosphere is absent, we want to make these measurements from the ground. However, it is difficult from the ground because it requires an adaptive optics system to provide an extremely well corrected wavefront to enable coronographic techniques. Currently only natural guide star AO systems have demonstrated the necessary wavefront correction for direct imaging of exoplanets. However, using a stellar source as the guide star for wavefront sensing limits the number of exoplanet systems we can directly image because it requires a relatively bright V~10 mag star. To increase the number of observable targets, we need to push the limit of natural guide stars to fainter magnitudes with high Strehl ratio correction. We propose to combine laser guide star (LGS) and natural guide star (NGS) wavefront sensing to achieve the high Strehl correction with fainter natural guide stars. We call this approach Hybrid Atmospheric Phase Analysis (HAPA); 'hapa' in Hawaiian means 'half' or 'of mixed ethnic heritage'. The relatively bright LGS is used for higher order correction, whereas the NGS is used for high accuracy lower order correction. We focus on demonstrating this approach using Robo-AO-2 at the UH 2.2m telescope on Maunakea with a UV Rayleigh laser at 355 nm. The laser focuses at 10 km altitude and has an equivalent magnitude of m_U~8. In this report specifically, we present simulated results of HAPA employed at Robo-AO-2, with the LGS system having a single configuration of 16x16 subaperture Shack-Hartmann wavefront sensor and the NGS system having 6 different configurations -- 16x16, 8x8, 5x5, 4x4, 2x2 and 1x1. We also discuss the on-sky experiments we plan to carry out with HAPA at the UH 2.2m telescope.

astro-ph.IM↗

arXiv2024

In-lab and On-sky Closed-loop Results of Adaptive Secondary Mirrors with TNO's Hybrid Variable Reluctance Actuators

We performed closed-loop lab testing of large-format deformable mirrors (DMs) with hybrid variable reluctance actuators. TNO has been developing the hybrid variable reluctance actuators in support for a new generation of adaptive secondary mirrors (ASMs), which aim to be more robust and reliable. Compared to the voice coil actuators, this new actuator technology has a higher current to force efficiency, and thus can support DMs with thicker facesheets. Before putting this new technology on-sky, it is necessary to understand how to control it and how it behaves in closed-loop. We performed closed-loop tests with the Shack-Hartmann wavefront sensor with three large-format deformable mirrors that use the TNO actuators: DM3, FLASH, and IRTF-ASM-1 ASM. The wavefront sensor and the real-time control systems were developed for the NASA Infrared Telescope Facility (IRTF) and the UH 2.2-meter telescope ASMs. We tested IRTF-ASM-1 on-sky and proved that it meets all of our performance requirements. This work presents our lab setup for the experiments, the techniques we have employed to drive these new ASMs, the results of our closed-loop lab tests for FLASH and IRTF-ASM-1, and the on-sky closed-loop results of IRTF-ASM-1 ASM.

astro-ph.IM↗

arXiv2024

First laboratory and on-sky results of an adaptive secondary mirror with TNO-style actuators on the NASA Infrared Telescope Facility

We are developing an adaptive secondary mirror (ASM) that uses a new actuator technology created by the Netherlands Organization for Applied Scientific Research (TNO). The TNO hybrid variable reluctance actuators have more than an order of magnitude better efficiency over the traditional voice coil actuators that have been used on existing ASMs and show potential for improving the long-term robustness and reliability of ASMs. To demonstrate the performance, operations, and serviceability of TNO's actuators in an observatory, we have developed a 36-actuator prototype ASM for the NASA Infrared Telescope Facility (IRTF) called IRTF-ASM-1. IRTF-ASM-1 provides the first on-sky demonstration of this approach and will help us evaluate the long-term performance and use of this technology in an astronomical facility environment. We present calibration and performance results with the ASM in a Meniscus Hindle Sphere lens setup as well as preliminary on-sky results on IRTF. IRTF-ASM-1 achieved stable closed-loop performance on-sky with H-band Strehl ratios of 35-40% in long-exposure images under a variety of seeing conditions.

astro-ph.IM↗

arXiv2024

Commissioning results from the Robo-AO-2 facility for rapid visible and near-infrared AO imaging

We installed the next-generation automated laser adaptive optics system, Robo-AO-2, on the University of Hawaii 2.2-m telescope on Maunakea in 2023. We engineered Robo-AO-2 to deliver robotic, diffraction-limited observations at visible and near-infrared wavelengths in unprecedented numbers. This new instrument takes advantage of upgraded components, manufacturing techniques and control; and includes a parallel reconfigurable natural guide star wavefront sensor with which to explore hybrid wavefront sensing techniques. We present the results of commissioning in 2023 and 2024.

astro-ph.IM↗

arXiv2024

Certified Robust Accuracy of Neural Networks Are Bounded due to Bayes Errors

Adversarial examples pose a security threat to many critical systems built on neural networks. While certified training improves robustness, it also decreases accuracy noticeably. Despite various proposals for addressing this issue, the significant accuracy drop remains. More importantly, it is not clear whether there is a certain fundamental limit on achieving robustness whilst maintaining accuracy. In this work, we offer a novel perspective based on Bayes errors. By adopting Bayes error to robustness analysis, we investigate the limit of certified robust accuracy, taking into account data distribution uncertainties. We first show that the accuracy inevitably decreases in the pursuit of robustness due to changed Bayes error in the altered data distribution. Subsequently, we establish an upper bound for certified robust accuracy, considering the distribution of individual classes and their boundaries. Our theoretical results are empirically evaluated on real-world datasets and are shown to be consistent with the limited success of existing certified training results, e.g., for CIFAR10, our analysis results in an upper bound (of certified robust accuracy) of 67.49\%, meanwhile existing approaches are only able to increase it from 53.89\% in 2017 to 62.84\% in 2023.

stat.ML↗