BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Meng Jin

Meng Jin contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
astro-ph.SRastro-ph.IMphysics.space-phArtificial Intelligenceastro-ph.EPDistributed, Parallel, and Cluster Computinghep-ph

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
8works
0followers
7topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

8 published item(s)

preprint2026arXiv

Safactory: A Scalable Agentic Infrastructure for Training Trustworthy Autonomous Intelligence

As large models evolve from conversational assistants into autonomous agents, challenges increasingly arise from long-horizon decision making, tool use, and real environment interaction. Existing agenticinfrastructure remain fragmented across evaluation, data management, and agent evolution, making it difficult to discover risks systematically and improve models in a continuous closed loop. In this report, we present \textbf{Safactory}, a scalable agent factory for trustworthy autonomous intelligence. Safactory integrates three tightly coupled platforms: a \textbf{Parallel Simulation Platform} for trajectory generation, a \textbf{Trustworthy Data Platform} for trajectory storage and experience extraction, and an \textbf{Autonomous Evolution Platform} for asynchronous reinforcement learning and on-policy distillation. As far as we know, Safactory is the first framework to propose a unified evolutionary pipeline for next-generation trustworthy autonomous intelligence.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Assessing the Influence of Input Magnetic Maps on Global Modeling of the Solar Wind and CME-driven Shock in the 2013 April 11 Event

In the past decade, significant efforts have been made in developing physics-based solar wind and coronal mass ejection (CME) models, which have been or are being transferred to national centers (e.g., SWPC, CCMC) to enable space weather predictive capability. However, the input data coverage for space weather forecasting is extremely limited. One major limitation is the solar magnetic field measurements, which are used to specify the inner boundary conditions of the global magnetohydrodynamic (MHD) models. In this study, using the Alfven wave solar model (AWSoM), we quantitatively assess the influence of the magnetic field map input (synoptic/diachronic vs. synchronic magnetic maps) on the global modeling of the solar wind and the CME-driven shock in the 2013 April 11 solar energetic particle (SEP) event. Our study shows that due to the inhomogeneous background solar wind and dynamical evolution of the CME, the CME-driven shock parameters change significantly both spatially and temporally as the CME propagates through the heliosphere. The input magnetic map has a great impact on the shock connectivity and shock properties in the global MHD simulation. Therefore this study illustrates the importance of taking into account the model uncertainty due to the imperfect magnetic field measurements when using the model to provide space weather predictions.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Coronal Mass Ejections and Dimmings: A Comparative Study using MHD Simulations and SDO Observations

Solar coronal dimmings have been observed extensively in the past two decades. Due to their close association with coronal mass ejections (CMEs), there is a critical need to improve our understanding of the physical processes that cause dimmings as well as their relationship with CMEs. In this study, we investigate coronal dimmings by combining simulation and observational efforts. By utilizing a data-constrained global magnetohydrodynamics model (AWSoM: Alfven-wave Solar Model), we simulate coronal dimmings resulting from different CME energetics and flux rope configurations. We synthesize the emissions of different EUV spectral bands/lines and compare with SDO/AIA and EVE observations. A detailed analysis of the simulation and observation data suggests that the transient dimming / brightening are related to plasma heating processes, while the long-lasting core and remote dimmings are caused by mass loss process induced by the CME. Moreover, the interaction between the erupting flux rope with different orientations and the global solar corona could significantly influence the coronal dimming patterns. Using metrics such as dimming depth and dimming slope, we investigate the relationship between dimmings and CME properties (e.g., CME mass, CME speed) in the simulation. Our result suggests that coronal dimmings encode important information about the associated CMEs, which provides a physical basis for detecting stellar CMEs from distant solar-like stars.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Mass Spectra and Decay of Mesons under Strong External Magnetic Field

We study the mass spectra and decay process of $σ$ and $π_0$ mesons under strong external magnetic field. For this purpose, we deduce the thermodynamic potential in a two-flavor, hot and magnetized Nambu-Jona-Lasinio model. We calculate the energy gap equation through the random phase approximation(RPA). Then we use Ritus method to calculate the decay triangle diagram and self-energy in the presence of a constant magnetic field B. Our results indicate that the magnetic field has little influence on the mass of $π_0$ at low temperatures. While for quarks and $σ$ mesons, their mass changes obviously, which reflects the influence of magnetic catalysis(MC). The presence of magnetic field accelerates the decay of the meson while the presence of chemical potential will decrease the decay process.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Simultaneous High Dynamic Range Algorithm, Testing, and Instrument Simulation

Within an imaging instrument's field of view, there may be many observational targets of interest. Similarly, within a spectrograph's bandpass, there may be many emission lines of interest. The brightness of these targets and lines can be orders of magnitude different, which poses a challenge to instrument and mission design. A single exposure can saturate the bright emission and/or have a low signal to noise ratio (SNR) for faint emission. Traditional high dynamic range (HDR) techniques solve this problem by either combining multiple sequential exposures of varied duration or splitting the light to different sensors. These methods, however, can result in the loss of science capability, reduced observational efficiency, or increased complexity and cost. The simultaneous HDR method described in this paper avoids these issues by utilizing a special type of detector whose rows can be read independently to define zones that are then composited, resulting in areas with short or long exposure measured simultaneously. We demonstrate this technique for the sun, which is bright on disk and faint off disk. We emulated these conditions in the lab to validate the method. We built an instrument simulator to demonstrate the method for a realistic solar imager and input. We then calculated SNRs, finding a value of 45 for a faint coronal mass ejection (CME) and 200 for a bright CME, both at 3.5 $R_{\odot}$ -- meeting or far exceeding the international standard for digital photography that defines a SNR of 10 as acceptable and 40 as excellent. Future missions should consider this type of hardware and technique in their trade studies for instrument design.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Probing the Physics of the Solar Atmosphere with the Multi-slit Solar Explorer (MUSE): II. Flares and Eruptions

Current state-of-the-art spectrographs cannot resolve the fundamental spatial (sub-arcseconds) and temporal scales (less than a few tens of seconds) of the coronal dynamics of solar flares and eruptive phenomena. The highest resolution coronal data to date are based on imaging, which is blind to many of the processes that drive coronal energetics and dynamics. As shown by IRIS for the low solar atmosphere, we need high-resolution spectroscopic measurements with simultaneous imaging to understand the dominant processes. In this paper: (1) we introduce the Multi-slit Solar Explorer (MUSE), a spaceborne observatory to fill this observational gap by providing high-cadence (<20 s), sub-arcsecond resolution spectroscopic rasters over an active region size of the solar transition region and corona; (2) using advanced numerical models, we demonstrate the unique diagnostic capabilities of MUSE for exploring solar coronal dynamics, and for constraining and discriminating models of solar flares and eruptions; (3) we discuss the key contributions MUSE would make in addressing the science objectives of the Next Generation Solar Physics Mission (NGSPM), and how MUSE, the high-throughput EUV Solar Telescope (EUVST) and the Daniel K Inouye Solar Telescope (and other ground-based observatories) can operate as a distributed implementation of the NGSPM. This is a companion paper to De Pontieu et al. (2021; arXiv:2106.15584), which focuses on investigating coronal heating with MUSE.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Atmospheric Escape From TOI-700 d: Venus versus Earth Analogs

The recent discovery of an Earth-sized planet (TOI-700 d) in the habitable zone of an early-type M-dwarf by the Transiting Exoplanet Survey Satellite constitutes an important advance. In this Letter, we assess the feasibility of this planet to retain an atmosphere -- one of the chief ingredients for surface habitability -- over long timescales by employing state-of-the-art magnetohydrodynamic models to simulate the stellar wind and the associated rates of atmospheric escape. We take two major factors into consideration, namely, the planetary atmospheric composition and magnetic field. In all cases, we determine that the atmospheric ion escape rates are potentially a few orders of magnitude higher than the inner Solar system planets, but TOI-700 d is nevertheless capable of retaining a $1$ bar atmosphere over gigayear timescales for certain regions of the parameter space. The simulations show that the unmagnetized TOI-700 d with a 1 bar Earth-like atmosphere could be stripped away rather quickly ($<$ 1 gigayear), while the unmagnetized TOI-700 d with a 1 bar CO$_2$-dominated atmosphere could persist for many billions of years; we find that the magnetized Earth-like case falls in between these two scenarios. We also discuss the prospects for detecting radio emission of the planet (thereby constraining its magnetic field) and discerning the presence of an atmosphere.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Predicting solar flares with machine learning: investigating solar cycle dependence

A deep learning network, Long-Short Term Memory (LSTM) network, is used in this work to predict whether the maximum flare class an active region (AR) will produce in the next 24 hours is class $Γ$. We considered $Γ$ are $\ge M$, $\ge C$ and any flare class. The essence of using LSTM, which is a recurrent neural network, is its capability to capture temporal information of the data samples. The input features are time sequences of 20 magnetic parameters from SHARPs - Space-weather HMI Active Region Patches. We analyzed active regions from June 2010 to Dec 2018, using the Geostationary Operational Environmental Satellite (GOES) X-ray flare catalogs and label the data samples with identified ARs in the GOES X-ray flare catalogs. Our results (i) shows consistent skill scores with recently published results using LSTMs and better than the previous work using single time input (eg. DeFN) (ii) The skill scores from the model show essential differences when different years of data was chosen for training and testing.

0 citations0 reviewsNo code linkedOpen access