arXiv subjects

Xiaoyang Chen

Publications and source records attributed to Xiaoyang Chen.

At least 19 recordsLinked to original sources

arXiv2026

A Wide and Deep Exploration of Radio-detected Active Galactic Nuclei with Subaru HSC (WERGS). XIII. High-z Radio Quasar Selection from HSC--VLASS over $\sim1200$ deg$^2$

We report the results of $g-$, $r-$, $i-$, and $z-$dropout selections based on optical identifications of Very Large Array Sky Survey (VLASS) radio sources using the Hyper Suprime-Cam Subaru Strategic Program survey (HSC--SSP). By positional crossmatching within $1.''5$ between the VLASS Epoch~2 catalog and the HSC--SSP Wide-layer catalog ($i \lesssim 26$), we obtain $\sim$400 high-redshift radio AGN candidates at $z \gtrsim 4$ over a $\approx1200~\mathrm{deg}^2$ survey footprint, extending optimistically to $z \sim 7$. Their $i$-band AB magnitudes range from $i_\mathrm{AB} \simeq 24$--26, placing most below the SDSS detection limit. By further crossmatching the HSC--VLASS dropout catalog with VLA Faint Images of the Radio Sky at Twenty-centimeters (FIRST) at 1.4~GHz, the LOFAR Two-metre Sky Survey (LoTSS) at 144~MHz, and the TIFR GMRT Sky Survey (TGSS) at 150~MHz, we obtain multi-frequency radio measurements for the majority of the sample. Approximately 85\% of the sources have three radio detections. The majority of the high-$z$ candidates show flat to moderately steep radio spectra ($-1 \lesssim α\lesssim 0$, with $f_ν\propto ν^α$), while some also exhibit turnover radio spectra, demonstrating a broad diversity of radio spectral properties among the dropout-selected candidates. SED fitting yields typical AGN bolometric luminosities of $\log(L_{\rm bol}/{\rm erg\,s^{-1}})\sim46.5$--47.5. We also find that the comoving number density declines from the $g$-dropout to the $z$-dropout regimes, with substantial decreases toward higher redshift, suggesting fewer luminous radio AGN candidates toward reionization.

astro-ph.GA↗

arXiv2026

Beyond Polarization: The Generative Constraint of Chain-of-Thought in Pointwise Reranking

In pointwise document reranking, Chain-of-Thought models typically underperform direct scoring models. While existing diagnostics attribute this to inferior classification, score polarization, or calibration breakdown, whether targeted training can bridge this gap remains unclear. Our empirical study first confirms that this gap is stable across scales up to 32B parameters, ruling out model and data capacity confounders. We then apply stress tests utilizing reinforcement learning, fine-grained supervision, and architectural decoupling to explicitly repair these deviations. Although these interventions improve classification accuracy and absolute scores, the relative ranking gap persists. These findings suggest that, within the pointwise scoring paradigm, routing continuous relevance semantics through discrete text constrains ranking signal resolution, revealing a bottleneck that is stable and difficult to overcome under current standard methods, rather than an easily resolvable training bias.

cs.CL↗

arXiv2026

DexMani: Human-Derived Manipulability Guidance for Dexterous Rotation

Dexterous object rotation is a sequential contact problem: each support, release, and re-contact decision must both produce the desired object motion, and prepare the hand configuration for continued rotation. Existing reinforcement learning methods discover such movement patterns through trial and error on specific robotic hand embodiments, without explicitly accounting for how each contact transition affects the hand's ability to sustain object rotation in subsequent steps. We introduce DexMani, a framework that transfers human demonstrations as contact-conditioned manipulability evolution. This prior captures how successful human contact transitions reshape the object-rotation directions available to the hand. DexMani then learns this manipulability evolution and uses it to guide downstream reinforcement learning, enabling rotation skills to be acquired across robot embodiments with distinct kinematics and active-contact configurations. Across the Shadow Hand, Allegro Hand, and XHand, DexMani achieves the highest success rates in every evaluated setting for both seen and unseen objects. DexMani reaches an average success rate of 57.5% on LEAP Hand, outperforming other baselines and producing smoother rotatory motions. Project site: https://dexmani.github.io

cs.RO↗

arXiv2026

NOVA: A Verification-Aware Agent Harness for Architecture Evolution in Industrial Recommender Systems

Industrial advertising recommender systems are continually improved through architecture modifications, yet production iteration remains expert-intensive because coordinated changes to model topology, feature configuration, and interaction modules must satisfy strict interface, resource, and serving constraints. AutoML is limited to predefined search spaces, while generic coding agents verify runnability rather than recommender-specific semantic validity. Executable candidates may therefore violate architectural contracts, while the lack of structured reuse of semantic diagnostics and evaluation outcomes can lead to repeated invalid or ineffective modifications. We present NOVA, a verification-aware agent harness that organizes production architecture modification as multi-round search over concrete implementations within a fixed evaluation budget. At each round, NOVA generates multiple candidates under production constraints, rejects semantic violations, and ranks the valid survivors for local testing and offline evaluation. Across rounds, trajectory memory synthesizes semantic diagnostics, local-test outcomes, and offline metric changes into modification directions and forbidden patterns that guide subsequent search. Under the same maximum offline-evaluation budget for automated methods, NOVA achieves the highest effective pass rate, reaching 53.3% on ScaleUp and 51.7% on Literature-to-Production tasks. In a production A/B test covering 5% of traffic in an advertising system serving over one billion users, the selected Literature-to-Production candidate yields GMV gains of +1.25%, +1.70%, and +2.02% across three major pCVR objectives, with corresponding relative reductions in absolute pCVR bias of 58.8%, 66.7%, and 37.3%, respectively.

cs.IR↗

arXiv2026

Rethinking Speech-LLM Integration for ASR: Effective Joint Speech-Text Training by Interleaving

Speech-LLM integration has shown promising results by leveraging extensive textual pretraining, yet its specific benefits for automatic speech recognition (ASR) remain unclear. We observe that as supervised ASR training data increases, the contribution of LLM priors becomes less evident, and simple speech-text joint training under-utilizes textual knowledge. We therefore propose Joint Speech-Text Interleaved Pretraining (JSTIP), an ASR-oriented pretraining strategy that constructs word-level and segment-level interleaved speech-text sequences within aligned pairs for speech-LLM architectures that accept continuous inputs. Experiments on 38k hours of ASR data show consistent entity accuracy improvement compared to ASR-only and joint speech-text training baselines. JSTIP achieves on-par entity recognition performance using domain transcription text compared to synthetic speech-text pairs, simplifying domain adaptation. Benefiting from textual pretraining and domain text data, JSTIP is competitive with open-source ASR and Speech-LLM systems in medical entity recognition. The zero-shot speech question answering behaviors further suggest that interleaving reduces the speech-text modality gap and preserves the LLM generative prior, which is likely the reason for the entity improvements on the ASR task.

cs.CL↗

arXiv2026

SpindleFlexNet: Flexible sleep spindles detection for EEG signals based on an adaptive one-dimensional RetinaNet-based framework

Sleep spindle is a physiologically significant biomedical signal in electroencephalographic (EEG) waveforms, which is typically a low-amplitude event in sleep. Due to the small signal ratio in the overall EEG, previous detection methods have limited capability to capture its start and end points and lack flexibility in handling multi-spindle scenarios. To address the gap, we address the problem from a new perspective and introduce SpindleFlexNet, the first framework in this field to apply deep learning-based one-dimensional object detection, leveraging an adapted one-dimensional RetinaNet architecture. The framework employs one-dimensional anchor generation, matching, and regression, along with a customized one-dimensional loss function. Analyses were conducted on two public datasets: the Montreal Archive of Sleep Studies and DREAMS, from which a total of 11,061 and 335 segments were obtained, respectively. When trained on these datasets, SpindleFlexNet achieved an average recall, precision, and F1-score of 0.61, 0.76, 0.67, and 0.58, 0.80, 0.67 in five-fold cross-validation. The model demonstrates stable detection performance and good generalization, making it a practical tool for sleep research. Potential applications include automated spindle labeling in clinical settings and as a reference for studies combining EEG with simultaneous functional magnetic resonance imaging.

eess.SP↗

arXiv2026

Towards Verifiable Agentic Data Science: Solving Irregular TSQA Via Tool-Grounded Reasoning

Time series data in real-world deployments is overwhelmingly irregular. Observations are asynchronous, missing values are informative rather than random, and sampling frequencies vary across sensors and operational windows. However, existing Time Series Question Answering (TSQA) benchmarks mostly assume regularly sampled inputs, leaving a fundamental gap in understanding how large language models (LLMs) and AI agents perform under irregular conditions. To bridge this gap, we introduce IRTS-ToolBench, a benchmark of 1,700 questions spanning 10 task types across 13 domains. IRTS-ToolBench is designed to be used independently by any researcher working on LLM-based irregular time series analysis, providing standardized inputs and a reproducible evaluation protocol. Code can be found in https://github.com/SanhornC/IRTS-ToolBench.

cs.AI↗

arXiv2026

Rank4Gen: RAG-Preference-Aligned Document Set Selection and Ranking

In the RAG paradigm, document ranking determines the evidence available to downstream generators. Through controlled analysis, we identify two phenomena underexplored by existing rankers: (i) downstream response quality depends not only on relevance but also on the composition and ordering of selected documents, and (ii) such preferences differ systematically across generators. However, existing rankers are trained purely on query--document relevance, leaving both phenomena unmodeled. To close this gap, we construct \textbf{PRISM}, a bilingual preference-aligned dataset built through a four-stage pipeline that compresses the combinatorial subset-and-ordering space by roughly four orders of magnitude and produces response-quality preference supervision conditioned on seven downstream generators. On a 13k-query subset of PRISM, we train \textbf{Rank4Gen}, a generator-aware ranker that performs joint document set selection and ordering. Experiments on five challenging RAG benchmarks show that Rank4Gen improves downstream QA quality on most evaluated generators, with per-generator F1 gains of up to $+2.08$ over the strongest set-selection baseline. Code is available at https://github.com/JOHNNY-fans/Rank4Gen.

cs.IR↗

arXiv2026

Moonshine: An Autonomous Mathematical Research Agent Centered on Conjecture Generation

Moonshine is an autonomous agent whose central objective is to generate mathematical conjectures. Its core capability is to extract structure from classical problems, distill new concepts, and formulate conjectures of mathematical significance. Rather than treating the solution of a single proposition as its endpoint, Moonshine builds an extensible theoretical framework through conjecture generation, bridge building, and obstacle identification. This article uses Moonshine's exploration of the Jacobian conjecture as an example. It shows how the central logic of whether local nondegeneracy can force global injectivity is transferred to one-hidden-layer affine-ridge sigmoid networks. This leads to the formulation of the \emph{Neural Jacobian Conjecture} (NJC): if such a network has strictly positive Jacobian determinant on the whole space, then it must be globally injective. By invoking GPT-5.5-pro and DeepSeek-V4-pro separately, Moonshine obtained independent complete proofs for the case $N=n+1$. In addition, with the assistance of ChatGPT through interactive use of its web interface with GPT-5.5-pro, a geometric-topological proof was developed. These results provide preliminary evidence for the plausibility of the conjecture. The general higher-width case $N\ge n+2$, however, remains unresolved and is left for further investigation. This work illustrates Moonshine's ability to autonomously generate meaningful mathematical problems and make rigorous progress on them.

cs.AI↗

arXiv2026

Revisiting Articulated Parts Perception in Robot Manipulation

We are surrounded by various objects with movable, articulated parts, e.g., box, handle, door. An accurate and generalizable perception of articulated parts is essential to enhance robotic manipulation capabilities. Building on this need, recent efforts in articulated parts perception have followed two main directions: One line of work uses pose-based representation, which requires high manual cost; in parallel, affordance-based methods extract future object motion from point tracking without additional manual efforts, but suffer from low-quality data. In this paper, we propose a new representation of articulated parts, Geometric Primary Structure (GPS), an abstraction of the part geometry structure to balance scalability and quality. For efficient and scalable data collection, GPS is integrated with a portable Virtual Reality (VR) device and requires only one minute to annotate one object sequence. This direct human annotation provides higher quality than the estimated affordance. With this efficient VR-GPS system, we collect 41K frames for 234 objects across six part classes, and train a generalizable GPS model with a single RGB-D object image as input. For object manipulation, we deploy a heuristic policy based on GPS prediction. Without any in-domain fine-tuning, our method achieves an 73% success rate, covering 270 initial states for 9 objects. Our code, data and reusable tool are available at https://enlighten0707.github.io/gps.

cs.RO↗

arXiv2026

A Wide and Deep Exploration of Radio-detected Active Galactic Nuclei with Subaru HSC (WERGS). XII. Final Optical Identification of VLASS Radio Sources from the Subaru/HSC-SSP Wide Survey Over 1200 deg$^2$

We present a wide-area and deep optical identification catalog for radio sources based on the VLASS Epoch 2 catalog at 3 GHz. Optical counterparts are identified using the final-year internal processing of the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP) Wide layer (DR S23B), which provides deep imaging over ~1200 deg$^2$ in $grizy$ with $i$-band depth of $i_\mathrm{AB}\sim26$. Starting from a 1.0 arcsec nearest-neighbor match between VLASS and HSC, we construct a quality-controlled primary catalog (Clean VLASS-HSC) of 22,773 sources by requiring S/N$>5$ in at least one HSC band. We further provide ancillary nearest-neighbor associations to FIRST (1.4 GHz) and LoTSS DR3 (150 MHz) within 2.5 arcsec, resulting in 18,444 FIRST-matched sources, 16,167 LoTSS-matched sources, and a 14,206-source subset matched to both surveys. The catalog contains approximately six times more optically identified radio sources than the first WERGS optical-identification effort based on the early HSC-SSP S16B data and positional cross-matching with FIRST (Yamashita et al. 2018). The VLASS Epoch 2 resolution yields more precise optical associations (median 0.199 arcsec) and, together with uniform FIRST/LoTSS matches, enables robust multi-frequency radio SED constraints. Compared to UNIONS-based VLASS identifications (Zhong et al. 2025), the deeper HSC imaging improves sensitivity to optically faint and even morphologically resolved hosts at $z\gtrsim1$. Our catalog preferentially highlights host-dominated AGN candidates, potentially including a substantial fraction of obscured systems.

astro-ph.GA↗

arXiv2026

Electronic Origin of Ferromagnetic Excitations in the Candidate Spin-Triplet Superconductor CeSb2

The origin of quasi-one-dimensional (q1D) ferromagnetic (FM) excitations in the candidate spin-triplet superconductor CeSb$_2$ has remained unclear. Here we report an electronic mechanism for emergent q1D magnetism in the quasi-two-dimensional lattice of CeSb$_2$, revealed by angle-resolved photoemission spectroscopy (ARPES). High-resolution ARPES resolves no spin-density-wave gap on the dispersive Fermi pockets, disfavoring a nesting-driven mechanism for the q1D FM excitations. Instead, resonant ARPES reveals a pronounced selective enhancement of Ce 4$f$ spectral weight on the $C_2$-distributed Fermi pockets aligned with the Ce ladder. This observation signifies band-selective Kondo coupling that generates strongly anisotropic magnetic exchange interactions, which can naturally account for both the q1D ferromagnetic excitations and the competing magnetic orders. Our results identify a band-selective Kondo coupling mechanism for emergent low-dimensional magnetism in correlated $f$-electron systems.

cond-mat.str-el↗

arXiv2026

$3d_{z^2}$ orbital delocalization and magnetic collapse in superconducting (La,Pr)$_3$Ni$_2$O$_{7-δ}$ films

The recent discovery of Ruddlesden--Popper (RP) nickelate thin-film superconductors has opened a new frontier in unconventional superconductivity. Its realization requires both compressive epitaxial strain and highly oxidative growth conditions, yet the microscopic pathway from the parent phase to the superconducting phase remains elusive. Here, X-ray absorption spectra and resonant inelastic X-ray scattering are employed to track this evolution by independently tuning strain and oxygen content in (La,Pr)$_3$Ni$_2$O$_{7-δ}$ thin films. We uncover a remarkable two-step narrative. First, signatures of delocalization emerge in the same way upon two independent tunings: Spectral weight transfers from a ''Upper Hubbard''-like peak to the hole-like peak associated with O $2p_z$ state, and in parallel, the initially localized Ni $3d_{z^2}$ orbital becomes more itinerant followed by the broadening and weakening of $dd$ orbital excitations. Second, as itinerancy increases, long-range spin-density-wave (SDW) order is suppressed in both intensity and correlation length, indicating direct competition with superconductivity. Yet, short-range magnons persist: they become damped but their bandwidth stays unchanged. Our results paint a coherent picture that both strain and oxygenation drive the RP bilayer nickelates towards the superconducting instability, where the O $2p_z$ and Ni $3d_{z^2}$ orbitals become delocalized. Concomitantly, the long-range magnetic order loses coherence and gets suppressed. These findings establish an orbital-selective route to RP nickelate superconductivity, in which the delocalization of the $2p_z$ and $3d_{z^2}$ orbitals and the robust short-range magnons upon the melting of SDW order are prerequisites, providing strong constraints for theory and the roadmap for designing nickelate superconductors.

cond-mat.supr-con↗

arXiv2026

Blackbody Quasar and Radio Source (BBQSORS): A Candidate of Transitional Little Red Dots with a $T\sim10^4\ K$ Blackbody Spectrum

We report Subaru/PFS spectroscopic follow-up of a radio-loud quasar at $z=1.715$ from the UNVEIL radio AGN catalog and with X-ray detections. The PFS spectrum displays a broad MgII emission line with an $\mathrm{FWHM}\gtrsim4000\ km/s$, accompanied by a narrow absorption feature. The spectrum reveals a characteristic $Λ$-shape over the rest-frame wavelength ranging $\sim1500-3500\ Å$. This underlying UV continuum is too curved to be reproduced by simply applying dust extinction to the spectrum of typical unobscured quasars. Alternatively, it is well described by a blackbody spectrum with a temperature of $T\approx10000\ K$. This result is in good agreement with its UV to MIR photometry that can be well modeled by three blackbody components representing the SMBH envelope ($\mathit{T}\approx9700\ K$), dust torus ($T\approx1500\ K$), and host galaxy dust ($T\approx80\ K$). The source is marginally detected in the GALEX NUV, revealing a potential V-shaped spectral energy distribution around $1400\ Å$, reminiscent of the spectral feature reported for recently discussed LRDs whose V-shapes occur around $3000-4000\ Å$. This wavelength shift is broadly consistent with the temperature contrast between our blackbody component, with $T\sim10^4\ K$, and the lower effective temperature of $T\sim5000\ K$ expected for an optically thick photosphere surrounding the SMBH in LRDs. These properties suggest that this source might be caught in a transient evolutionary phase in which the dense gas envelope characteristic of LRD has begun to fragment, allowing us to witness the emergence of a quasar from an LRD-like state.

astro-ph.GA↗

arXiv2026

Lectures on AI for Mathematics

This book provides a comprehensive and accessible introduction to the emerging field of AI for mathematics. It covers the core principles and diverse applications of using artificial intelligence to advance mathematical research. Through clear explanations, the text explores how AI can discover hidden mathematical patterns, assist in proving complicated theorems, and even construct counterexamples to challenge conjectures.

cs.AI↗

arXiv2026

Dissecting superconductivity in the Ruddlesden-Popper nickelates: The role of electron correlation and interlayer magnetic exchange

The discovery of superconductivity in the Ruddlesden-Popper (RP) nickelates has opened a new chapter in the search for high superconducting transition temperatures ($T_\mathrm{c}$) materials. A central and puzzling feature of this family is the wide variation in $T_\mathrm{c}$ despite their common NiO$_2$ building blocks, as highlighted by the recent observation of superconductivity at $\sim$ 30 K in trilayer $\mathrm{La_4Ni_3O_{10}}$, significantly lower than 80 K reported in bilayer $\mathrm{La_3Ni_2O_7}$. Understanding the factors that control $T_\mathrm{c}$ in this family is therefore of paramount importance. Here, we use resonant inelastic x-ray scattering (RIXS) to investigate the electronic and magnetic excitations of $\mathrm{La_4Ni_3O_{10}}$ in direct comparison with its bilayer counterpart. Our results reveal a markedly different landscape. $\mathrm{La_4Ni_3O_{10}}$ exhibits a more itinerant character, evidenced by broader Ni $dd$ orbital excitations and a strong Ni 3$d$ fluorescence continuum, suggesting weaker electronic correlations than in the bilayer. Despite this, well-defined collective spin excitations persist, including dispersive acoustic and optical magnon branches alongside an incommensurate spin density wave. Using linear spin wave theory, we extract the interlayer superexchange interaction ($J_z$) to be $\sim$ 22 meV, much smaller than that in $\mathrm{La_3Ni_2O_7}$. The weaker correlation and reduced interlayer exchange together provide a consistent explanation for the substantially lower $T_\mathrm{c}$ in the trilayer compound. Our findings establish interlayer magnetic coupling and electronic correlation as key parameters governing superconductivity in layered nickelates and offer critical constraints for understanding the pairing mechanism in this emerging family.

cond-mat.supr-con↗

arXiv2026

Can LLM generate interesting mathematical research problems?

This paper is the second one in a series of work on the mathematical creativity of LLM. In the first paper, the authors proposed three criteria for evaluating the mathematical creativity of LLM and constructed a benchmark dataset to measure it. This paper further explores the mathematical creativity of LLM, with a focus on investigating whether LLM can generate valuable and cutting-edge mathematical research problems. We develop an agent to generate unknown problems and produced 665 research problems in differential geometry. Through human verification, we find that many of these mathematical problems are unknown to experts and possess unique research value.

cs.AI↗

arXiv2026

A z ~ 0.4 Galaxy Reflecting the high-redshift Little Red Dots: An Extended Starburst with an Overmassive Black Hole

One of the most remarkable discoveries of JWST is a population of compact, red sources at z > 4, commonly referred to as Little Red Dots (LRDs). Spectroscopic identifications reported that most LRDs are active galactic nuclei (AGNs), which are preferentially found around z ~ 6 and could imply a key phase in the formation and growth of black holes (BHs) in the early universe. Photometric surveys at lower redshift have recently been carried out to trace their evolution across cosmic time, and a small number of LRD-like galaxies have been spectroscopically identified at both Cosmic Noon and in the local universe. Here we report the discovery of one of the lowest-z LRD-like galaxies, J204837.26-002437.2 (hereafter J2048) at z = 0.4330, using new Gemini-N/GMOS IFU observations combined with archival multi-band photometric SED data. The GMOS data reveal extended blue emission from starburst with a star formation rate of 400 solar mass per year, together with an extended, highly fast ionized outflow. This is the first spectroscopic confirmation of extended host emission and outflow in an LRD-like galaxy, providing a unique laboratory for understanding the nature of their high-redshift counterparts. Moreover, J2048 would host an extremely overmassive BH with a BH-to-stellar mass ratio of ~ 60%, with the BH mass and host stellar mass estimated to be 10^10.2 and 10^10.4 solar masses, respectively. We discuss the origin and evolutionary fate of J2048, and the implications that such low-z analogs have for interpreting the properties of high-z LRDs.

astro-ph.GA↗