arXiv subjects

Jiyuan Zhang

Publications and source records attributed to Jiyuan Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2026

A Newly Identified Degeneracy Keeps the Planetary Interpretation Viable for OGLE-2011-BLG-0950

The microlensing event OGLE-2011-BLG-0950 exhibits the well-known ``Planet/Binary'' degeneracy, in which distinct lens configurations produce similar light curves but imply substantially different mass ratios between the lens components. A previous study suggested that high-resolution imaging could break this degeneracy through differences in the lens-source relative proper motion. In this work, we identify a new planetary model for this event that arises from a newly identified degeneracy, simultaneously reproducing the observed light curve and remaining consistent with the relative proper motion measured from high-resolution imaging. By combining constraints from the light-curve modeling and high-resolution observations, we infer a lens system consisting of a $\sim 1~M_{\odot}$ host star orbited by a $\sim 1.5~M_{\rm Jup}$ planet, with a projected separation of about 2 or 8 au, subject to the ``Close/Wide'' degeneracy. Our reanalysis of the color-magnitude diagram further indicates that the source star has unresolved companions that contribute non-negligible blended light, highlighting the importance of carefully accounting for source and lens companions in future Roman microlensing analyses. Finally, we show that adopting a single mass--luminosity relation significantly underestimates the uncertainties in the inferred lens properties for host masses $\gtrsim 1~M_{\odot}$.

astro-ph.EP↗

arXiv2026

KMT-2025-BLG-1616Lb: First Microlensing Bound Planet From DREAMS

We present observations and analysis of the bound planetary microlensing event KMT-2025-BLG-1616. The planetary signal was captured by the Korea Microlensing Telescope Network (KMTNet) and the DECam Rogue Earths and Mars Survey (DREAMS). DREAMS's minute-cadence observations break the central/resonant degeneracy in the binary-lens models. The color of the faint source star ($I=22$) is measured from the DREAMS's $r - z$ color. The planetary system has a planet-host mass ratio of $q \sim 5 \times 10^{-4}$. A Bayesian analysis yields a host-star mass of $\sim 0.3\,M_\odot$, a planetary mass of $\sim 40\,M_{\oplus}$, a projected planet-host separation of $\sim 1.6~\mathrm{au}$, and a lens distance of $\sim 7.5~\mathrm{kpc}$. Based on the photometric precision achieved by DREAMS for this event, we simulate free-floating planet (FFP) detections and find that DREAMS is sensitive to Mars-mass FFPs in the Galactic bulge and Moon-mass FFPs in the Galactic disk.

astro-ph.EP↗

arXiv2026

MoEBlaze: Breaking the Memory Wall for Efficient MoE Training on Modern GPUs

The pervasive "memory wall" bottleneck is significantly amplified in modern large-scale Mixture-of-Experts (MoE) architectures. MoE's inherent architectural sparsity leads to sparse arithmetic compute and also introduces substantial activation memory overheads -- driven by large token routing buffers and the need to materialize and buffer intermediate tensors. This memory pressure limits the maximum batch size and sequence length that can fit on GPUs, and also results in excessive data movements that hinders performance and efficient model scaling. We present MoEBlaze, a memory-efficient MoE training framework that addresses these issues through a co-designed system approach: (i) an end-to-end token dispatch and MoE training method with optimized data structures to eliminate intermediate buffers and activation materializing, and (ii) co-designed kernels with smart activation checkpoint to mitigate memory footprint while simultaneously achieving better performance. We demonstrate that MoEBlaze can achieve over 4x speedups and over 50% memory savings compared to existing MoE frameworks.

cs.LG↗

arXiv2025

Matrix harmonic analysis at high temperature via the Dirichlet process

We investigate harmonic analysis of random matrices of large size with their Dyson indices going simultaneous to zero, that is in the high temperature limit. In this regime, we show that the multivariate Bessel function/Heckman-Opdam hypergeometric function of the empirical spectral distribution converges to the Fourier/Mellin transform of a measure, which and the limiting empirical distribution are intimately related by the Markov-Krein correspondence. The uniqueness, existence and other properties of the Markov-Krein correspondence can be studied using the theory of the Dirichlet process.

math-ph↗

arXiv2025

Four Giant Planets from 2024 KMTNet Microlensing Campaign

In this work, we present analyses of four newly discovered planetary microlensing events from the 2024 KMTNet survey season: KMT-2024-BLG-0176, KMT-2024-BLG-0349, KMT-2024-BLG-1870, and KMT-2024-BLG-2087. In each case, the planetary nature was revealed through distinct types of anomalies in the lensing light curves: a positive bump near the peak for KMT-2024-BLG-0176, an asymmetric peak for KMT-2024-BLG-0349, a short-duration central dip for KMT-2024-BLG-1870, and a caustic-crossing feature for KMT-2024-BLG-2087. Detailed modeling of the light curves confirms that these anomalies are produced by planetary companions with planet-to-host mass ratios in the range of $(1.5\text{--}17.9)\times 10^{-3}$. Despite the diversity of signal morphologies, all planets detected in these events are giant planets with masses comparable to or exceeding that of Jupiter in the Solar System. Each planet orbits a host star less massive than the Sun, emphasizing the strength of microlensing in uncovering planetary systems around low-mass stellar hosts.

astro-ph.EP↗

arXiv2025

OmniSparse: Training-Aware Fine-Grained Sparse Attention for Long-Video MLLMs

Existing sparse attention methods primarily target inference-time acceleration by selecting critical tokens under predefined sparsity patterns. However, they often fail to bridge the training-inference gap and lack the capacity for fine-grained token selection across multiple dimensions such as queries, key-values (KV), and heads, leading to suboptimal performance and limited acceleration gains. In this paper, we introduce OmniSparse, a training-aware fine-grained sparse attention framework for long-video MLLMs, which operates in both training and inference with dynamic token budget allocation. Specifically, OmniSparse contains three adaptive and complementary mechanisms: (1) query selection via lazy-active classification, retaining active queries that capture broad semantic similarity while discarding most lazy ones that focus on limited local context and exhibit high functional redundancy; (2) KV selection with head-level dynamic budget allocation, where a shared budget is determined based on the flattest head and applied uniformly across all heads to ensure attention recall; and (3) KV cache slimming to reduce head-level redundancy by selectively fetching visual KV cache according to the head-level decoding query pattern. Experimental results show that OmniSparse matches the performance of full attention while achieving up to 2.7x speedup during prefill and 2.4x memory reduction during decoding.

cs.CV↗

arXiv2025

A Comprehensive Analysis of Three Microlensing Planet Candidates with the Planet/Binary Degeneracy

We present observations and analyses of three high-magnification microlensing events: KMT-2022-BLG-0954, KMT-2024-BLG-0697, and MOA-2024-BLG-018. All three exhibit the "Planet/Binary" degeneracy, with planetary solutions corresponding to mass ratios in the range $-3.7 < \log q < -2.2$, while the binary solutions yield $\log q > -2.0$. For KMT-2022-BLG-0954, we identify a previously unrecognized degeneracy among planetary solutions, involving different mass ratios and normalized source radii. In all three cases, single-lens binary-source models are excluded. Bayesian analyses suggest that the planetary solutions correspond to gas giants orbiting M/K dwarfs beyond the snow line, while KMT-2022-BLG-0954 also admits an alternative interpretation as a super-Earth orbiting a late-type M dwarf. The binary solutions imply a diverse set of systems, including M-dwarf pairs and M-dwarf--brown-dwarf binaries. A review of known events subject to the "Planet/Binary" degeneracy shows that in most cases the degeneracy cannot be resolved through follow-up high-resolution imaging, particularly in the presence of the newly identified degeneracy.

astro-ph.EP↗

arXiv2025

Exploring Nonlinear System with Machine Learning: Chua and Lorenz Circuits Analyzed

Nonlinear circuits serve as crucial carriers and physical models for investigating nonlinear dynamics and chaotic behavior, particularly in the simulation of biological neurons. In this study, Chua's circuit and Lorenz circuit are systematically explored for the first time through machine learning correlation algorithms. Specifically, the upgraded and optimized SINDy-PI model, which is based on neural network and symbolic regression algorithm, is utilized to learn the numerical results of attractors generated by these two nonlinear circuits. Through error analysis, we examine the effects of the precision of input data and the amount of data on the algorithmic model. The findings reveal that when the input data quantity and data precision fall within a certain range, the algorithm model can effectively recognize and restore the differential equation expressions corresponding to the two circuits. Additionally, we test the anti-interference ability of different circuits and the robustness of the algorithm by introducing noise into the test data. The results indicate that under the same noise disturbance, the Lorenz circuit has better noise resistance than Chua's circuit, providing a starting point for further studying the intrinsic properties and characteristics of different nonlinear circuits. The above results will not only offer a reference for the further study of nonlinear circuits and related systems using deep learning algorithms but also lay a preliminary theoretical foundation for the study of related physical problems and applications.

physics.comp-ph↗

arXiv2025

USP-Gaussian: Unifying Spike-based Image Reconstruction, Pose Correction and Gaussian Splatting

Spike cameras, as an innovative neuromorphic camera that captures scenes with the 0-1 bit stream at 40 kHz, are increasingly employed for the 3D reconstruction task via Neural Radiance Fields (NeRF) or 3D Gaussian Splatting (3DGS). Previous spike-based 3D reconstruction approaches often employ a casecased pipeline: starting with high-quality image reconstruction from spike streams based on established spike-to-image reconstruction algorithms, then progressing to camera pose estimation and 3D reconstruction. However, this cascaded approach suffers from substantial cumulative errors, where quality limitations of initial image reconstructions negatively impact pose estimation, ultimately degrading the fidelity of the 3D reconstruction. To address these issues, we propose a synergistic optimization framework, \textbf{USP-Gaussian}, that unifies spike-based image reconstruction, pose correction, and Gaussian splatting into an end-to-end framework. Leveraging the multi-view consistency afforded by 3DGS and the motion capture capability of the spike camera, our framework enables a joint iterative optimization that seamlessly integrates information between the spike-to-image network and 3DGS. Experiments on synthetic datasets with accurate poses demonstrate that our method surpasses previous approaches by effectively eliminating cascading errors. Moreover, we integrate pose optimization to achieve robust 3D reconstruction in real-world scenarios with inaccurate initial poses, outperforming alternative methods by effectively reducing noise and preserving fine texture details. Our code, data and trained models will be available at https://github.com/chenkang455/USP-Gaussian.

cs.CV↗

arXiv2025

Systematic Search for FFPs in KMTNet Full-Frame Images. I. Photometry Pipeline

To exhume the buried signatures of free-floating planets (FFPs) with small angular Einstein radius $θ_{\rm E}$, we build a new full-frame difference image pipeline for the Korean Microlensing Telescope Network (KMTNet) survey based on the newly optimized pySIS package. We introduce the detailed processes of the new pipeline, including frame registration, difference image analysis, and light curve extraction. To test this pipeline, we extract 1-year light curves for 483,068 stars with $I \lesssim 17$ and conduct a model-independent search for microlensing events. The search finds 36 microlensing events, including five new events and six events discovered by other collaborations but missed by previous KMTNet searches. We find that the light curves from the new pipeline are precise enough to be sensitive to FFPs with $θ_{\rm E} \sim 1~μ$as. Using the new pipeline, a complete FFP search on the eight-year KMTNet images can be finished within six months and then yield the FFP mass function. The new pipeline can be used for a new KMTNet AlertFinder system, with significantly reduced false positives.

astro-ph.EP↗

arXiv2025

Evaluating and Advancing Multimodal Large Language Models in Perception Ability Lens

As multimodal large language models (MLLMs) advance rapidly, rigorous evaluation has become essential, providing further guidance for their development. In this work, we focus on a unified and robust evaluation of \textbf{vision perception} abilities, the foundational skill of MLLMs. We find that existing perception benchmarks, each focusing on different question types, domains, and evaluation metrics, introduce significant evaluation variance, complicating comprehensive assessments of perception abilities when relying on any single benchmark. To address this, we introduce \textbf{AbilityLens}, a unified benchmark designed to evaluate MLLMs in six key perception abilities (ranging from counting, OCR, to understanding structural data), focusing on both accuracy and stability, with each ability encompassing diverse types of questions, domains, and metrics. With the assistance of AbilityLens, we: (1) identify the strengths and weaknesses of current main-stream MLLMs, highlighting stability patterns and revealing a notable performance gap between state-of-the-art open-source and closed-source models; (2) uncover interesting ability conflict and early convergence phenomena during MLLM training; (3) reveal the primary reason of ability conflict is data mixing ratio and LLM model size; and (4) discuss the effectiveness of some straightforward strategies \eg, fine-tuning and model merging, to solve the ability conflict. The benchmark and online leaderboard is released in https://github.com/Chenfeng1271/AbilityLens.

cs.CV↗

arXiv2025

X-Driver: Explainable Autonomous Driving with Vision-Language Models

End-to-end autonomous driving has advanced significantly, offering benefits such as system simplicity and stronger driving performance in both open-loop and closed-loop settings than conventional pipelines. However, existing frameworks still suffer from low success rates in closed-loop evaluations, highlighting their limitations in real-world deployment. In this paper, we introduce X-Driver, a unified multi-modal large language models(MLLMs) framework designed for closed-loop autonomous driving, leveraging Chain-of-Thought(CoT) and autoregressive modeling to enhance perception and decision-making. We validate X-Driver across multiple autonomous driving tasks using public benchmarks in CARLA simulation environment, including Bench2Drive[6]. Our experimental results demonstrate superior closed-loop performance, surpassing the current state-of-the-art(SOTA) while improving the interpretability of driving decisions. These findings underscore the importance of structured reasoning in end-to-end driving and establish X-Driver as a strong baseline for future research in closed-loop autonomous driving.

cs.RO↗

arXiv2025

KMT-2022-BLG-1818Lb,c: A Cold Super-Jupiter with a Saturn Sibling

We present the discovery and analysis of the sixth microlensing two-planet system, KMT-2022-BLG-1818Lb,c, detected by a follow-up program targeting high-magnification events. Both planets are subject to the well-known ''Close/Wide'' degeneracy, although for the first planet, which has a super-Jovian mass ratio of $q_2 \simeq 5\times 10^{-3}$ in both solutions, the Close topology, with a normalized separation of $s\simeq 0.70$, is clearly preferred by $Δχ^2=26$. However, contrary to all previous two-planet microlensing systems, the mass ratio for the second planet, $q_3$, is substantially (factor of $\sim 10$) different for the Close and Wide topologies of the first planet. While this degeneracy is resolved in the present case due to high-cadence follow-up observations, the appearance of this new degeneracy indicates the need for caution in the analysis of future two-planet systems. A Bayesian analysis suggests that the host is likely a K-dwarf star in the Galactic disk. The first planet is probably a super-Jupiter on a Jupiter-like orbit, while the second planet is a Saturn-class planet on either a Mercury-like or Saturn-like orbit.

astro-ph.EP↗

arXiv2025

Microlensing events indicate that super-Earth exoplanets are common in Jupiter-like orbits

Exoplanets classified as super-Earths are commonly observed on short period orbits, close to their host stars, but their abundance on wider orbits is poorly constrained. Gravitational microlensing is sensitive to exoplanets on wide orbits. We observed the microlensing event OGLE-2016-BLG-0007, which indicates an exoplanet with a planet-to-star mass ratio roughly double the Earth-Sun mass-ratio, on an orbit longer than Saturn's. We combine this event with a larger sample from a microlensing survey to determine the distribution of mass ratios for planets on wide orbits. We infer there are $\sim 0.35$ super-Earth planets per star on Jupiter-like orbits. The observations are most consistent with a bimodal distribution, with separate peaks for super-Earths and gas giants. We suggest that this reflects differences in their formation processes.

astro-ph.EP↗

arXiv2025

Systematic KMTNet Planetary Anomaly Search. XII. Complete Sample of 2017 Subprime Field Planets

We report the analysis of four unambiguous planets and one possible planet from the subprime fields ($Γ\leq 1~{\rm hr}^{-1}$) of the 2017 Korea Microlensing Telescope Network (KMTNet) microlensing survey, to complete the KMTNet AnomalyFinder planetary sample for the 2017 subprime fields. They are KMT-2017-BLG-0849, KMT-2017-BLG-1057, OGLE-2017-BLG-0364, and KMT-2017-BLG-2331 (unambiguous), as well as KMT-2017-BLG-0958 (possible). For the four unambiguous planets, the mean planet-host mass ratios, $q$, are $(1.0, 1.2, 4.6, 13) \times 10^{-4}$, the median planetary masses are $(6.4, 24, 76, 171)~M_{\oplus}$ and the median host masses are $(0.19, 0.57, 0.49, 0.40)~M_{\odot}$ from a Bayesian analysis. We have completed the AnomalyFinder planetary sample from the first 4-year KMTNet data (2016--2019), with 112 unambiguous planets in total, which nearly tripled the microlensing planetary sample. The ``sub-Saturn desert'' ($\log q = \left[-3.6, -3.0\right]$) found in the 2018 and 2019 KMTNet samples is confirmed by the 2016 and 2017 KMTNet samples.

astro-ph.EP↗

arXiv2025

Systematic Reanalysis of KMTNet Microlensing Events, Paper II: Two New Planets in Giant-Source Events

In this work, we continue to apply the updated KMTNet tender-love care (TLC) photometric pipeline to historical microlensing events. We apply the pipeline to a subsample of events from the KMTNet database, which we refer to as the giant source sample. Leveraging the improved photometric data, we conduct a systematic search for anomalies within this sample. The search successfully uncovers four new planet-like anomalies and recovers two previously known planetary signals. After detailed analysis, two of the newly discovered anomalies are confirmed as clear planets: KMT-2019-BLG-0578 and KMT-2021-BLG-0736. Their planet-to-host mass ratios are $q\sim4\times10^{-3}$ and $q\sim1\times10^{-4}$, respectively. Another event, OGLE-2018-BLG-0421 (KMT-2018-BLG-0831), remains ambiguous. Both a stellar companion and a giant planet in the lens system could potentially explain the observed anomaly. The anomaly signal of the last event, MOA-2022-BLG-038 (KMT-2022-BLG-2342), is attributed to an extra source star. Within this sample, our procedure doubles the number of confirmed planets, demonstrating a significant enhancement in the survey sensitivity.

astro-ph.EP↗

arXiv2025

Exploring the Role of Explicit Temporal Modeling in Multimodal Large Language Models for Video Understanding

Applying Multimodal Large Language Models (MLLMs) to video understanding presents significant challenges due to the need to model temporal relations across frames. Existing approaches adopt either implicit temporal modeling, relying solely on the LLM decoder, or explicit temporal modeling, employing auxiliary temporal encoders. To investigate this debate between the two paradigms, we propose the Stackable Temporal Encoder (STE). STE enables flexible explicit temporal modeling with adjustable temporal receptive fields and token compression ratios. Using STE, we systematically compare implicit and explicit temporal modeling across dimensions such as overall performance, token compression effectiveness, and temporal-specific understanding. We also explore STE's design considerations and broader impacts as a plug-in module and in image modalities. Our findings emphasize the critical role of explicit temporal modeling, providing actionable insights to advance video MLLMs.

cs.CV↗

arXiv2024

Stable distributions and domains of attraction for unitarily invariant Hermitian random matrix ensembles

We consider random matrix ensembles on the set of Hermitian matrices that are heavy tailed, in particular not all moments exist, and that are invariant under the conjugate action of the unitary group. The latter property entails that the eigenvectors are Haar distributed and, therefore, factorise from the eigenvalue statistics. We prove a classification for stable matrix ensembles of this kind of matrices represented in terms of matrices, their eigenvalues and their diagonal entries with the help of the classification of the multivariate stable distributions and the harmonic analysis on symmetric matrix spaces. Moreover, we identify sufficient and necessary conditions for their domains of attraction. To illustrate our findings we discuss for instance elliptical invariant random matrix ensembles and Pólya ensembles, the latter playing a particular role in matrix convolutions. As a byproduct we generalise the derivative principle on the Hermitian matrices to general tempered distributions. This principle relates the joint probability density of the eigenvalues and the diagonal entries of the random matrix.

math.PR↗