arXiv subjects

Su Yao

Publications and source records attributed to Su Yao.

At least 19 recordsLinked to original sources

arXiv2026

A New Sample of $\sim$ 100 Intermediate-mass Black Holes Reaching $z \approx 1$

We present a systematic search for intermediate-mass black hole (IMBH) active galactic nuclei (AGNs) at $0.5 < z \lesssim 1$ using DESI DR1 spectroscopy.We identify 98 broad-line IMBH AGNs with black hole masses $M_\mathrm{BH}<10^6$ $M_{\odot}$ through quantitative spectral decomposition and broad-H$\beta$ selection. This sample spans $M_\mathrm{BH}=10^{5.5}-10^{6.0}$ $M_{\odot}$ and Eddington ratios from 1.2 to 9.0 extending systematic IMBH AGN searches to intermediate redshift. Compared with a consistently selected $z<0.6$ IMBH sample, the $0.5<z<1$ sources reach higher broad-H$\beta$ luminosities and substantially higher Eddington ratios, indicating more extreme accretion states at earlier cosmic times. They also show broader [O III] profiles and stronger blueshifted wing components, with these kinematic differences persisting after matching in black hole mass and Eddington ratio. These results reveal two distinct signatures of evolution among IMBH AGNs at $z<1$: ability of IMBHs to reach increasingly extreme accretion states toward higher redshift, and systematic changes in their ionized-gas kinematics. The former demonstrates that rapid, including super-Eddington, growth of IMBHs can persist to relatively late cosmic times, while the latter may indicate evolution in the ionized-gas environment and associated outflow activity of actively growing IMBHs. Together, these findings provide new constraints on the evolutionary pathways of IMBHs at $z<1$, and show that seed-mass black holes can continue to undergo rapid growth well after the cosmic dawn.

astro-ph.GA↗

arXiv2026

Comparative analysis of missing data imputation methods for CSST survey: Impact on photometric redshift estimation performance

Improving the accuracy of photometric redshifts (photo-$z$) is essential for reliable statistical studies of cosmology and galaxy evolution. However, missing photometric bands are a common observational challenge that can significantly degrade photo-$z$ estimation accuracy. In this work, we present a systematic evaluation of data imputation methods aimed at improving photo-$z$ performance. We benchmark a range of representative machine learning (ML) and deep learning (DL) architectures, identifying k-nearest neighbors (KNN) and the attention-based SAITS model as the leading performers. These models are then applied to China Space Station Survey Telescope (CSST) mock data to assess their performance under realistic observational conditions. Our results show that KNN yields the highest accuracy under idealized missing completely at random (MCAR) conditions with complete training sets, whereas robustness tests reveal that SAITS significantly outperforms KNN when training data is incomplete or when applied to realistic mixed-mechanism scenarios. We find that domain consistency between training and testing missingness patterns is a prerequisite for optimal performance, highlighting the risks of domain shift in supervised regression tasks. Furthermore, our analysis demonstrates that while general imputation models are highly effective for MCAR and missing at random (MAR) data, they are detrimental when applied to missing not at random (MNAR) data arising from flux limits, as statistical models fail to capture the physical information inherent in these non-detections. Consequently, we advocate for more sophisticated architectures capable of disentangling stochastic missingness from physical non-detections to address these distinct mechanisms individually.

astro-ph.GA↗

arXiv2026

RPS: Information Elicitation with Reinforcement Prompt Selection

Large language models (LLMs) have shown remarkable capabilities in dialogue generation and reasoning, yet their effectiveness in eliciting user-known but concealed information in open-ended conversations remains limited. In many interactive AI applications, such as personal assistants, tutoring systems, and legal or clinical support, users often withhold sensitive or uncertain information due to privacy concerns, ambiguity, or social hesitation. This makes it challenging for LLMs to gather complete and contextually relevant inputs. In this work, we define the problem of information elicitation in open-ended dialogue settings and propose Reinforcement Prompt Selection (RPS), a lightweight reinforcement learning framework that formulates prompt selection as a sequential decision-making problem. To analyze this problem in a controlled setting, we design a synthetic experiment, where a reinforcement learning agent outperforms a random query baseline, illustrating the potential of policy-based approaches for adaptive information elicitation. Building on this insight, RPS learns a policy over a pool of prompts to adaptively elicit concealed or incompletely expressed information from users through dialogue. We also introduce IELegal, a new benchmark dataset constructed from real legal case documents, which simulates dialogue-based information elicitation tasks aimed at uncovering case-relevant facts. In this setting, RPS outperforms static prompt baselines, demonstrating the effectiveness of adaptive prompt selection for eliciting critical information in LLM-driven dialogue systems.

cs.LG↗

arXiv2026

A Robust Geometric Distortion Solution for Main Survey Camera of CSST

The advancement in sensitivity and field of view of next-generation wide-field survey telescopes requires astrometric measurements with high precision, even in the presence of significant geometric distortions. To address this challenge, we develop a Weighted Polynomial Distortion Correction in 2-Phase (WPDC-2P) method. This approach enhances stellar cross-matching, incorporates distance-based weighting into the traditional polynomial fitting, and employs a look-up table to absorb the remaining distortion residuals. Validated on simulated data from the Main Survey Camera of the \emph{Chinese Space Station Survey Telescope} (CSST), incorporating geometric distortions up to approximately $200$ pixels, the method achieves astrometric standard deviation ranging from 0.013 to 0.107 pixels (0.03 pixels for the $g$-1 detector) across all 18 detectors. Under extreme crowding conditions (e.g., globular cluster NGC 2298), the astrometric precision for the $g$-1 detector reaches 0.05-pixel level within the central region ($r_d < 4000$), despite a centroiding precision of $\sim$0.04 pixels. When applied to the Beijing-Arizona Sky Survey data, for which the standard pipeline delivers an astrometric uncertainty of $\sim$20 mas, our method reduces the positional scatter to $ \sigma_{\Delta\alpha}=5.494$ mas (0.01 pixels) and $ \sigma_{\Delta\delta}=9.981$ mas (0.02 pixels) using only a weighted 3rd-order polynomial correction. The method has been integrated into the CSST data processing pipeline and is prepared for further refinement using on-orbit calibration data.

astro-ph.IM↗

arXiv2026

Selecting Optimal Stellar Calibration Fields for the CSST Imaging Survey

The Chinese Space Station Survey Telescope (CSST) will perform a decade-long high-precision wide-field imaging survey that relies on rigorous on-orbit calibration. This necessitates stable celestial benchmark fields to maintain photometric and astrometric consistency throughout the mission lifetime. We establish comprehensive selection criteria including observational visibility, stellar number density, bright-star contamination, and interstellar dust extinction. Using the CSST Observation Strategy Analysis Tool (COSAT) and all-sky dust maps from Planck and SFD, we constrain eligible regions to the ranges of ecliptic latitude $ |\beta| > 50^\circ$ and galactic latitude $|b| > 15^\circ$. From an initial sample of 29 candidate clusters meeting these spatial constraints, six globular clusters (M13, M92, NGC 104, NGC 362, NGC 1261, and NGC 1851) are identified as optimal calibration fields, fulfilling all the critical criteria. These selected clusters are recommended as optimal calibration field candidates for CSST's on-orbit calibration program, and are fundamental to achieving unprecedented photometric precision in CSST's space-based survey.

astro-ph.SR↗

arXiv2026

Transferable Graph Condensation from the Causal Perspective

The increasing scale of graph datasets has significantly improved the performance of graph representation learning methods, but it has also introduced substantial training challenges. Graph dataset condensation techniques have emerged to compress large datasets into smaller yet information-rich datasets, while maintaining similar test performance. However, these methods strictly require downstream applications to match the original dataset and task, which often fails in cross-task and cross-domain scenarios. To address these challenges, we propose a novel causal-invariance-based and transferable graph dataset condensation method, named TGCC, providing effective and transferable condensed datasets. Specifically, to preserve domain-invariant knowledge, we first extract domain causal-invariant features from the spatial domain of the graph using causal interventions. Then, to fully capture the structural and feature information of the original graph, we perform enhanced condensation operations. Finally, through spectral-domain enhanced contrastive learning, we inject the causal-invariant features into the condensed graph, ensuring that the compressed graph retains the causal information of the original graph. Experimental results on five public datasets and our novel FinReport dataset demonstrate that TGCC achieves up to a 13.41% improvement in cross-task and cross-domain complex scenarios compared to existing methods, and achieves state-of-the-art performance on 5 out of 6 datasets in the single dataset and task scenario.

cs.LG↗

arXiv2026

When Helpers Become Hazards: A Benchmark for Analyzing Multimodal LLM-Powered Safety in Daily Life

As Multimodal Large Language Models (MLLMs) become an indispensable assistant in human life, the unsafe content generated by MLLMs poses a danger to human behavior, perpetually overhanging human society like a sword of Damocles. To investigate and evaluate the safety impact of MLLMs responses on human behavior in daily life, we introduce SaLAD, a multimodal safety benchmark which contains 2,013 real-world image-text samples across 10 common categories, with a balanced design covering both unsafe scenarios and cases of oversensitivity. It emphasizes realistic risk exposure, authentic visual inputs, and fine-grained cross-modal reasoning, ensuring that safety risks cannot be inferred from text alone. We further propose a safety-warning-based evaluation framework that encourages models to provide clear and informative safety warnings, rather than generic refusals. Results on 18 MLLMs demonstrate that the top-performing models achieve a safe response rate of only 57.2% on unsafe queries. Moreover, even popular safety alignment methods limit effectiveness of the models in our scenario, revealing the vulnerabilities of current MLLMs in identifying dangerous behaviors in daily life. Our dataset is available at https://github.com/xinyuelou/SaLAD.

cs.CL↗

arXiv2025

The Large Sky Area Multi-object Fiber Spectroscopic Telescope (LAMOST) Quasar Survey: Quasar Properties from Data Release 10 to 12

We present the quasar catalog from Data Releases 10 to 12 of the Large Sky Area Multi-Object Fiber Spectroscopic Telescope (LAMOST) Quasar Survey, comprising quasars observed between September 2021 and June 2024. We robustly identified $11,346$ quasars, of which $5,386$ are newly discovered objects not present in the Million Quasars catalog. This release brings the total number of quasars identified by the 12-year LAMOST survey to $67,521$, of which $29,513$ are newly discovered. While the absolute flux calibration for LAMOST quasar spectra from Data Releases 6 to 9 was previously performed using the SDSS/PanSTARRS1 multi-band photometric data, the inherent variability of quasars can affect the flux accuracy. To address this limitation, we recalibrated the LAMOST spectra using (quasi-)simultaneous photometric data from Zwicky Transient Facility (ZTF), which has conducted high-cadence sky monitoring since March 2018. Based on the recalibrated single-epoch spectra, we estimated the emission line fluxes, continuum fluxes, and virial black hole masses. These improved spectra facilitate direct comparison with the spectra of common quasars from the Sloan Digital Sky Survey (SDSS), enabling searches for rare quasars, such as changing-look quasars exhibiting the appearance or disappearance of broad emission lines and broad absorption line quasars. The combined dataset of photometry and multi-epoch spectra will enhance the detections of AGN-related transients, such as Bowen fluorescence flares and extreme variability quasars, thereby improving our understanding of quasar variability.

astro-ph.GA↗

arXiv2025

FENIX: Enabling In-Network DNN Inference with FPGA-Enhanced Programmable Switches

Machine learning (ML) is increasingly used in network data planes for advanced traffic analysis, but existing solutions (such as FlowLens, N3IC, BoS) still struggle to simultaneously achieve low latency, high throughput, and high accuracy. To address these challenges, we present FENIX, a hybrid in-network ML system that performs feature extraction on programmable switch ASICs and deep neural network inference on FPGAs. FENIX introduces a Data Engine that leverages a probabilistic token bucket algorithm to control the sending rate of feature streams, effectively addressing the throughput gap between programmable switch ASICs and FPGAs. In addition, FENIX designs a Model Engine to enable high-accuracy deep neural network inference in the network, overcoming the difficulty of deploying complex models on resource-constrained switch chips. We implement FENIX on a programmable switch platform that integrates a Tofino ASIC and a ZU19EG FPGA directly, and evaluate it on real-world network traffic datasets. Our results show that FENIX achieves microsecond-level inference latency and multi-terabit throughput with low hardware overhead, and delivers over 90% accuracy on mainstream network traffic classification tasks, outperforming the state of the art.

cs.NI↗

arXiv2025

Pegasus: A Universal Framework for Scalable Deep Learning Inference on the Dataplane

The paradigm of Intelligent DataPlane (IDP) embeds deep learning (DL) models on the network dataplane to enable intelligent traffic analysis at line-speed. However, the current use of the match-action table (MAT) abstraction on the dataplane is misaligned with DL inference, leading to several key limitations, including accuracy degradation, limited scale, and lack of generality. This paper proposes Pegasus to address these limitations. Pegasus translates DL operations into three dataplane-oriented primitives to achieve generality: Partition, Map, and SumReduce. Specifically, Partition "divides" high-dimensional features into multiple low-dimensional vectors, making them more suitable for the dataplane; Map "conquers" computations on the low-dimensional vectors in parallel with the technique of fuzzy matching, while SumReduce "combines" the computation results. Additionally, Pegasus employs Primitive Fusion to merge computations, improving scalability. Finally, Pegasus adopts full precision weights with fixed-point activations to improve accuracy. Our implementation on a P4 switch demonstrates that Pegasus can effectively support various types of DL models, including Multi-Layer Perceptron (MLP), Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), and AutoEncoder models on the dataplane. Meanwhile, Pegasus outperforms state-of-the-art approaches with an average accuracy improvement of up to 22.8%, along with up to 248x larger model size and 212x larger input scale.

cs.NI↗

arXiv2025

A Sample of Active Galactic Nuclei with Intermediate-mass Black Holes Extended to $z \approx$ 0.6

We present a sample of 930 intermediate-mass black hole (IMBH) active galactic nuclei (AGNs) with black hole masses of $M_\mathrm{BH} \leqslant 2 \times 10^{6}$ M$_{\odot}$, uniformly selected from the Seventeenth Data Release of the Sloan Digital Sky Survey, based on the detection of broad H$\alpha$ or H$\beta$ emission lines. Taking advantage of the wide wavelength coverage of BOSS/eBOSS spectroscopy, our sample extends the redshift coverage of low-$z$ IMBH AGNs to $z\leqslant0.57$, significantly improving upon previous studies that where generally limited to $z\leqslant0.35$. This sample encompasses black hole masses from $10^{4.0}$ to $10^{6.3}$ M$_{\odot}$, with Eddington ratios ranging from 0.01 to 1.9. Among the $z>0.3$ subset, 24 sources exhibit detectable broad Mg II $\lambda\lambda2796,2803$ emission lines, including eight confirmed by independent DESI spectra. A preliminary analysis reveals a marked decline in both the maximum accretion rate ($L/L_\mathrm{Edd}$) and broad H$\alpha$ luminosity with decreasing redshift, possibly reflecting a cosmic evolution of accretion activity at the low-mass end, akin to the ``downsizing'' evolutionary trend seen in high-mass AGNs.

astro-ph.GA↗

arXiv2025

X-ray Emission Properties of a Compact Symmetric Object Sample

We present a comprehensive analysis of the X-ray observations obtained from \xmm\, and \chandra\, for a sample of bona-fide Compact Symmetric Objects (CSOs) to investigate their X-ray emission properties. Ultimately, we obtain 32 effective X-ray observational spectra from 17 CSOs. Most spectra can be well described by an absorbed single power-law model, with the exception of 6 spectra requiring an additional component in the soft X-ray band and 2 spectra exhibiting an iron emission line component. The data analysis results unveil the diverse characteristics of X-ray emission from CSOs. The sample covers X-ray luminosity ranging within $10^{40}-10^{45}$ erg s$^{-1}$, intrinsic absorbing column density ($N_{\rm H}^{\rm int}$) ranging within $10^{20}-10^{23}$ cm$^{-2}$, and photon spectral index ($\Gamma_{\rm X}$) ranging within 0.75--3.0. None of the CSOs in our sample have $N_{\rm H}^{\rm int}$ > $10^{23}\rm~cm^{-2}$, indicating that the X-ray emission in these CSOs is not highly obscured. The distribution of $\Gamma_{\rm X}$ for these CSOs closely resembles that observed in a sample of radio-loud quasars and low-excitation radio galaxies (RGs). In the radio--X-ray luminosity panel, these CSOs exhibit a distribution more akin to FR I RGs than FR II RGs, characterized by higher luminosities. The positive correlation between $\Gamma_{\rm X}$ and the Eddington ratio, which has been noted in radio-quiet active galactic nuclei, is not observed in these CSOs. These findings suggest that although the contribution of the disk-corona system cannot be completely ruled out, jet/lobe radiation likely plays a dominant role in the X-ray emission of these CSOs.

astro-ph.HE↗

arXiv2024

First deep X-ray observations of the Fermi-detected steep-spectrum source and radio-loud NLS1 galaxy 3C 286

A well-known calibrator source in radio astronomy, 3C 286 ($z=0.85$), is a compact steep-spectrum (CSS) radio source and spectroscopically classified as a narrow-line Seyfert 1 (NLS1) galaxy. It is also known for its damped Ly$\alpha$ system from an intervening galaxy at $z=0.692$ detected in both ultraviolet (UV) and radio spectra. In addition, despite being a misaligned active galactic nuclei (AGN), 3C 286 is also detected in $\gamma$-rays by Fermi. Thus, this unique object combines the characteristics of CSS sources, NLS1 galaxies, and $\gamma$-ray emitters with misaligned jets, providing an excellent laboratory for extending our knowledge of AGN disk-jet coupling. Despite its significance, 3C 286 has been rarely observed in X-rays. In this study, we present our deep XMM-Newton and Chandra observations of 3C 286. The results reveal that the X-ray spectrum can be well described by models including an intervening absorber with redshift and column density consistent with previous UV and radio observations. The most important finding is that the spectrum cannot be described by a single power law, but a soft excess is required which is parameterized by a blackbody. Furthermore, we find evidence suggesting the presence of off-nuclear X-ray emission at a radius that corresponds to the location of the radio lobes. While further theoretical work is still needed, our findings offer new clues to understand the specific mechanism for $\gamma$-ray emission from this unique object.

astro-ph.HE↗

arXiv2024

Walking in Others' Shoes: How Perspective-Taking Guides Large Language Models in Reducing Toxicity and Bias

The common toxicity and societal bias in contents generated by large language models (LLMs) necessitate strategies to reduce harm. Present solutions often demand white-box access to the model or substantial training, which is impractical for cutting-edge commercial LLMs. Moreover, prevailing prompting methods depend on external tool feedback and fail to simultaneously lessen toxicity and bias. Motivated by social psychology principles, we propose a novel strategy named \textbf{perspective-taking prompting (\textsc{PeT})} that inspires LLMs to integrate diverse human perspectives and self-regulate their responses. This self-correction mechanism can significantly diminish toxicity (up to $89\%$) and bias (up to $73\%$) in LLMs' responses. Rigorous evaluations and ablation studies are conducted on two commercial LLMs (ChatGPT and GLM) and three open-source LLMs, revealing \textsc{PeT}'s superiority in producing less harmful responses, outperforming five strong baselines.

cs.CL↗

arXiv2024

Magnetically arrested disks in FR I radio galaxies

A sample of 17 FR I radio galaxies constructed from the 3CR catalog, which is characterized by edge-darkened radio structures, is studied. The optical core luminosities derived from Hubble Space Telescope observation are used to estimate the Eddington ratios which are found to be below $10^{-3.4}$ for this sample. This is supported by the Baldwin-Phillips-Terlevich optical diagnostic diagrams derived with the spectroscopic observation of Telescopio Nazionale Galileo, suggesting that these sources are of low ionization nuclear Emission-line Regions (LINERs). It implies that the accretion in these FR I sources can be modeled as advection-dominated accretion flows (ADAFs). Given the low accretion rate, the predicted jet power with a fast-spinning black hole (BH) $a=0.95$ in the Blandford-Znajek mechanics is lower than the estimated one for almost all the sources in our sample. Such powerful jets indicate the presence of magnetically arrested disks (MAD) in the inner region of the ADAF, in the sense that the magnetic fields in the inner accretion zone are strong. Moreover, we show that, even in the MAD scenario, the BH spins in the sample are most likely moderate and/or fast with $a\gtrsim0.5$.

astro-ph.CO↗

arXiv2023

Multiwavelength variability of gamma-ray emitting narrow-line Seyfert 1 galaxies

As one of the drivers of feedback in active galactic nuclei (AGNs), the jets launched from supermassive black holes (SMBHs) are important for understanding the co-evolution of SMBHs and their host galaxies. However, the formation of AGN jets is far from clear. The discovery of gamma-ray narrow-line Seyfert 1 (NLS1) galaxies during the past two decades has provided us with a new means of studying the link between jets and accretion processes and the formation of jets. Here, we explore the coupling of jet and accretion discs in seven bright gamma-ray NLS1 galaxies by studying simultaneous optical/ultraviolet and X-ray observations of these systems taken by Swift. The results show that, except for 1H 0323+342 in which the X-rays are significantly contributed from the accretion disc, the observed X-ray emission of the other sources is dominated by the jet, and accretion process makes little contribution if not absent. Although the origin of the X-ray emission is different, the broad-band spectral shape characterized by alpha_ox and the X-ray flux is found to follow the same evolutionary trend in 1H 0323+342, PMN J0948+0022, and PKS 1502+036. For the remaining sources, the trend is not observed or the sampling is not dense enough.

astro-ph.HE↗

arXiv2023

Unveiling the small-scale jets in the rapidly growing supermassive black hole IZw1

Accretion of black holes at near-Eddington or super-Eddington rates is the most powerful episode that drives black hole growth, and it may work in several types of objects. However, the physics of accretion and jet-disc coupling in such a state remains unclear, mainly because the associated jets are not easily detectable due to the extremely weak emission or possibly episodic nature of the jets. Only a few near/super-Eddington systems have demonstrated radio activity, and it remains unclear whether there is a jet and what are their properties, in super-Eddington active galactic nuclei (AGNs) (and ultraluminous X-ray sources). The deficit is mainly due to the complex radio mixing between the origins of jets and others, such as star formation activity, photo-ionized gas, accretion disk wind, and coronal activity. In this work, we conducted high-resolution very long baseline interferometry (VLBI) observations to explore the jets in the highly accreting narrow-line Seyfert I system IZw1. Our observations successfully revealed small-scale jets (with a linear size of $\sim45$ parsec) at both 1.5 and 5 GHz, based on the high radio brightness temperature, radio morphology, and spectral index distribution. Interestingly, the lack of a flat-spectrum radio core and knotty jet structures imply episodic ejections in IZw1, which resemble the ejection process in Galactic X-ray binaries that are in the canonical very high state. The high accretion rates and jet properties in the AGN IZw1 may support the AGN/XRB analogy in the extreme state.

astro-ph.HE↗

arXiv2023

Differentially Private Learning with Per-Sample Adaptive Clipping

Privacy in AI remains a topic that draws attention from researchers and the general public in recent years. As one way to implement privacy-preserving AI, differentially private learning is a framework that enables AI models to use differential privacy (DP). To achieve DP in the learning process, existing algorithms typically limit the magnitude of gradients with a constant clipping, which requires carefully tuned due to its significant impact on model performance. As a solution to this issue, latest works NSGD and Auto-S innovatively propose to use normalization instead of clipping to avoid hyperparameter tuning. However, normalization-based approaches like NSGD and Auto-S rely on a monotonic weight function, which imposes excessive weight on small gradient samples and introduces extra deviation to the update. In this paper, we propose a Differentially Private Per-Sample Adaptive Clipping (DP-PSAC) algorithm based on a non-monotonic adaptive weight function, which guarantees privacy without the typical hyperparameter tuning process of using a constant clipping while significantly reducing the deviation between the update and true batch-averaged gradient. We provide a rigorous theoretical convergence analysis and show that with convergence rate at the same order, the proposed algorithm achieves a lower non-vanishing bound, which is maintained over training iterations, compared with NSGD/Auto-S. In addition, through extensive experimental evaluation, we show that DP-PSAC outperforms or matches the state-of-the-art methods on multiple main-stream vision and language tasks.

cs.LG↗