Searcharxiv⌕ Search

arXiv subjects

Jing Li

Publications and source records attributed to Jing Li.

At least 469 records · Page 26Linked to original sources
arXiv2023

Giant interfacial spin-Hall angle from Rashba-Edelstein effect revealed by the spin-Hall Hanle processes

The Rashba-Edelstein effect (REE), which generates interfacial spin polarization and subsequent spin current, is a compelling spin-charge conversion mechanism for spintronics applications, since it is not limited by the elemental spin-orbit coupling. In this work, we demonstrate REE at Pt/ferroelectric interfaces using the recently elucidated spin-Hall Hanle effects (SHHE), in which a Larmor precession of spin polarization in a diffusion process from the interface manifest as magnetoresistance and Hall effect. We show that REE leads to a three-fold enhancement of the effective spin Hall angle in ferroelectric interface Pt/h-LuFeO3 compared to that of Pt /Al2O3, although the difference in the spin relaxation time is negligible. Modeling using SHHEs involving REE as an additional source of interfacial polarization suggests that REE can lead to an interfacial spin Hall angle (~0.3) that is one order of magnitude larger than the bulk value of Pt. Our results demonstrate that a ferroelectric interface can produce large spin-charge conversion and that SHHEs are a sensitive tool for characterizing interfacial spin transport properties.

cond-mat.mes-hall↗
arXiv2023

FedLALR: Client-Specific Adaptive Learning Rates Achieve Linear Speedup for Non-IID Data

Federated learning is an emerging distributed machine learning method, enables a large number of clients to train a model without exchanging their local data. The time cost of communication is an essential bottleneck in federated learning, especially for training large-scale deep neural networks. Some communication-efficient federated learning methods, such as FedAvg and FedAdam, share the same learning rate across different clients. But they are not efficient when data is heterogeneous. To maximize the performance of optimization methods, the main challenge is how to adjust the learning rate without hurting the convergence. In this paper, we propose a heterogeneous local variant of AMSGrad, named FedLALR, in which each client adjusts its learning rate based on local historical gradient squares and synchronized learning rates. Theoretical analysis shows that our client-specified auto-tuned learning rate scheduling can converge and achieve linear speedup with respect to the number of clients, which enables promising scalability in federated optimization. We also empirically compare our method with several communication-efficient federated optimization methods. Extensive experimental results on Computer Vision (CV) tasks and Natural Language Processing (NLP) task show the efficacy of our proposed FedLALR method and also coincides with our theoretical findings.

cs.LG↗
arXiv2023

Physical Properties of the Young Asteroid Pair 2010 UM26 and 2010 RN221

The main belt asteroids 458271 (2010 UM26) and 2010 RN221 share almost identical orbital elements and currently appear as comoving objects 30 arcsec apart in the plane of the sky. They are products of the breakup of a parent object, or the splitting of a binary, with a separation age measured in decades rather than thousands or millions of years as for most other asteroid pairs (Vokrouhlicky et al.~2022). The nature of the precursor body and the details of the breakup and separation of the components are unknown. We obtained deep, high resolution imaging using the Hubble Space Telescope to characterize the pair and to search for material in addition to the main components that might have been released upon breakup. The primary and secondary have absolute magnitudes $H$ = 17.98 and 19.69, respectively, and effective diameters 760 m and 350 m (assuming geometric albedo 0.20). The secondary/primary mass ratio is 0.1, assuming equal densities. Time-series photometry shows that the primary rotates with period 5.9 hour and has a small photometric range (0.15 magnitudes), while the period of the secondary is undetermined (but >20 hours) and its lightcurve range is at least 1 magnitude. The primary rotation period and component mass ratio are consistent with a simple model for the breakup of a rotationally unstable precursor. However, unlike other observationally supported instances of asteroid breakup, neither macroscopic fragments nor unresolved material are found remaining in the vicinity of this asteroid pair. We suggest that the pair is a recently dissociated binary, itself formed earlier by rotational instability of 2010 UM26.

astro-ph.EP↗
arXiv2023

Tunable Chiral Bound States in a Dimer Chain of Coupled Resonators

We study the chiral feature in a system composed of one two-level emitter (TLE) and a one dimensional (1D) dimer chain of coupled resonators with the alternate single-photon energies. In the single-excitation subspace, this system not only possesses two energy bands with propagating states, but also possesses chiral bound states. The number of chiral bound states depends on the coupling forms between the TLE and the dimer chain. It is found that when the TLE is locally coupled to one resonator of the dimer chain, the bound-state that has mirror reflection symmetry is not a chiral one. When the TLE is nonlocally coupled to two adjacent resonators, three chiral bound states arise due to the mirror symmetry breaking. The chirality of these bound states can be tuned by changing the energy differences of single photon in the adjacent resonators, the coupling strengths and the transition energy of the TLE. The chirality behaviour of the ordinary two bound states outside the energy bands is quite different from the one of the emerging bound state inside the energy gap. The almost perfect chiral bound states can be achieved at certain parameters as a result of completely destructive interference.

physics.optics↗
arXiv2023

HICL: Hashtag-Driven In-Context Learning for Social Media Natural Language Understanding

Natural language understanding (NLU) is integral to various social media applications. However, existing NLU models rely heavily on context for semantic learning, resulting in compromised performance when faced with short and noisy social media content. To address this issue, we leverage in-context learning (ICL), wherein language models learn to make inferences by conditioning on a handful of demonstrations to enrich the context and propose a novel hashtag-driven in-context learning (HICL) framework. Concretely, we pre-train a model #Encoder, which employs #hashtags (user-annotated topic labels) to drive BERT-based pre-training through contrastive learning. Our objective here is to enable #Encoder to gain the ability to incorporate topic-related semantic information, which allows it to retrieve topic-related posts to enrich contexts and enhance social media NLU with noisy contexts. To further integrate the retrieved context with the source text, we employ a gradient-based method to identify trigger terms useful in fusing information from both sources. For empirical studies, we collected 45M tweets to set up an in-context NLU benchmark, and the experimental results on seven downstream tasks show that HICL substantially advances the previous state-of-the-art results. Furthermore, we conducted extensive analyzes and found that: (1) combining source input with a top-retrieved post from #Encoder is more effective than using semantically similar posts; (2) trigger words can largely benefit in merging context from the source and retrieved posts.

cs.CL↗
arXiv2023

Safety in Traffic Management Systems: A Comprehensive Survey

Traffic management systems play a vital role in ensuring safe and efficient transportation on roads. However, the use of advanced technologies in traffic management systems has introduced new safety challenges. Therefore, it is important to ensure the safety of these systems to prevent accidents and minimize their impact on road users. In this survey, we provide a comprehensive review of the literature on safety in traffic management systems. Specifically, we discuss the different safety issues that arise in traffic management systems, the current state of research on safety in these systems, and the techniques and methods proposed to ensure the safety of these systems. We also identify the limitations of the existing research and suggest future research directions.

eess.SY↗
arXiv2023

Helping students apply superposition principle in problems involving spherical, cylindrical and planar charge distributions

We describe student difficulties in applying the superposition principle in combination with Gauss's law. We addressed these difficulties by developing a tutorial that uses guided inquiry. Students who used this tutorial following lecture-based instruction performed significantly better on these topics than those who did not. Instructors can assign the tutorial as classwork or homework.

physics.ed-ph↗
arXiv2023

Earning Extra Performance from Restrictive Feedbacks

Many machine learning applications encounter a situation where model providers are required to further refine the previously trained model so as to gratify the specific need of local users. This problem is reduced to the standard model tuning paradigm if the target data is permissibly fed to the model. However, it is rather difficult in a wide range of practical cases where target data is not shared with model providers but commonly some evaluations about the model are accessible. In this paper, we formally set up a challenge named \emph{Earning eXtra PerformancE from restriCTive feEDdbacks} (EXPECTED) to describe this form of model tuning problems. Concretely, EXPECTED admits a model provider to access the operational performance of the candidate model multiple times via feedback from a local user (or a group of users). The goal of the model provider is to eventually deliver a satisfactory model to the local user(s) by utilizing the feedbacks. Unlike existing model tuning methods where the target data is always ready for calculating model gradients, the model providers in EXPECTED only see some feedbacks which could be as simple as scalars, such as inference accuracy or usage rate. To enable tuning in this restrictive circumstance, we propose to characterize the geometry of the model performance with regard to model parameters through exploring the parameters' distribution. In particular, for the deep models whose parameters distribute across multiple layers, a more query-efficient algorithm is further tailor-designed that conducts layerwise tuning with more attention to those layers which pay off better. Extensive experiments on different applications demonstrate that our work forges a sound solution to the EXPECTED problem. Code is available via https://github.com/kylejingli/EXPECTED.

cs.LG↗
arXiv2023

The Dimorphos Boulder Swarm

We present deep Hubble Space Telescope images taken to examine the ejecta from the DART spacecraft impact into asteroid Dimorphos. The images reveal an extensive population of co-moving boulders, the largest of which is about 7 m in diameter (geometric albedo 0.15 assumed). Measurements of 37 boulders show a mean sky-plane velocity dispersion of 0.30+/-0.03 m/s, only slightly larger than the 0.24 m/s gravitational escape velocity from the Didymos/Dimorphos binary system. The total boulder mass, 5e6 kg (density 2200 kg/m3 assumed), corresponds to about 0.1 percent of the mass of Dimorphos and the boulders collectively carry about 3e-5 of the kinetic energy delivered by the DART spacecraft impact. The sky-plane distribution of the boulders is asymmetric, consistent with impact into an inhomogeneous, likely rubble-pile, body. Surface boulder counts on Didymos show that the observed boulder swarm could be ejected from as little as 2 percent of the surface of Dimorphos (for example a circular crater at the impact point about 50 m in diameter). The large, slow-moving boulders are potential targets to be investigated in-situ by the upcoming ESA HERA mission.

astro-ph.EP↗
arXiv2023

Capacity Expansion of High Renewable Penetrated Energy Systems Considering Concentrating Solar Power for Seasonal Energy Balance

With the increasing proportion of variable renewable energy which owns fluctuation characteristics and the promotion of the Clean Heating policy, the seasonal energy imbalance of the system has been more and more challenging. There is a lack of effective means to mitigate this challenge under the background of gradual compression of the traditional thermal unit construction. Concentrating solar power (CSP) is a promising technology to replace thermal units by integrating emergency boilers to cope with extreme weather, and can meet long-time energy balance as a seasonal peak regulation source. In this paper, we propose a long-term high-resolution expansion planning model of the energy system under high renewable penetration which integrates CSP technology for seasonal energy balance. With the projection to 2050, by taking the energy system in Xinjiang province which is a typical area of the Clean Heating project with rich irradiance as a case study, it shows that the optimal deployment of CSP and electric boiler (EB) can reduce the cost, peak-valley difference of net load and renewable curtailment by 8.73%, 19.72% and 58.24% respectively at 65% renewable penetration compared to the base scenario.

eess.SY↗
arXiv2023

Electronic Polarization Effects in Core-Level Spectroscopy

In X-ray photoelectron spectroscopy (XPS), the injected hole interacts with the electronic polarization cloud induced by the hole itself, ultimately resulting in a lower binding energy. Such polarization effect can shift the core-level energy by more than 1 eV, as shown here by embedded many-body perturbation theory for the paradigmatic case of noble gas clusters made of Ar, Kr, or Xe. The polarization energy is almost identical for the different core-orbitals of a given atom, but it strongly depends on the position of the ionized atom in the cluster. An analytical formula is derived from classical continuum electrostatics, providing an effective and accurate description of polarization effects, which permits to achieve an excellent agreement with available experiments on noble gas clusters at a modest computational cost. Electronic polarization provides a crucial contribution to core levels absolute energies and chemical shifts.

cond-mat.mes-hall↗
arXiv2023

Rethinking Document-Level Relation Extraction: A Reality Check

Recently, numerous efforts have continued to push up performance boundaries of document-level relation extraction (DocRE) and have claimed significant progress in DocRE. In this paper, we do not aim at proposing a novel model for DocRE. Instead, we take a closer look at the field to see if these performance gains are actually true. By taking a comprehensive literature review and a thorough examination of popular DocRE datasets, we find that these performance gains are achieved upon a strong or even untenable assumption in common: all named entities are perfectly localized, normalized, and typed in advance. Next, we construct four types of entity mention attacks to examine the robustness of typical DocRE models by behavioral probing. We also have a close check on model usability in a more realistic setting. Our findings reveal that most of current DocRE models are vulnerable to entity mention attacks and difficult to be deployed in real-world end-user NLP applications. Our study calls more attentions for future research to stop simplifying problem setups, and to model DocRE in the wild rather than in an unrealistic Utopian world.

cs.CL↗
arXiv2023

Gate reflectometry in dense quantum dot arrays

Silicon quantum devices are maturing from academic single- and two-qubit devices to industrially-fabricated dense quantum-dot (QD) arrays, increasing operational complexity and the need for better pulsed-gate and readout techniques. We perform gate-voltage pulsing and gate-based reflectometry measurements on a dense 2$\times$2 array of silicon quantum dots fabricated in a 300-mm-wafer foundry. Utilizing the strong capacitive couplings within the array, it is sufficient to monitor only one gate electrode via high-frequency reflectometry to establish single-electron occupation in each of the four dots and to detect single-electron movements with high bandwidth. A global top-gate electrode adjusts the overall tunneling times, while linear combinations of side-gate voltages yield detailed charge stability diagrams. To test for spin physics and Pauli spin blockade at finite magnetic fields, we implement symmetric gate-voltage pulses that directly reveal bidirectional interdot charge relaxation as a function of the detuning between two dots. Charge sensing within the array can be established without the involvement of adjacent electron reservoirs, important for scaling such split-gate devices towards longer 2$\times$N arrays. Our techniques may find use in the scaling of few-dot spin-qubit devices to large-scale quantum processors.

cond-mat.mes-hall↗
arXiv2023

Topic-Guided Self-Introduction Generation for Social Media Users

Millions of users are active on social media. To allow users to better showcase themselves and network with others, we explore the auto-generation of social media self-introduction, a short sentence outlining a user's personal interests. While most prior work profiles users with tags (e.g., ages), we investigate sentence-level self-introductions to provide a more natural and engaging way for users to know each other. Here we exploit a user's tweeting history to generate their self-introduction. The task is non-trivial because the history content may be lengthy, noisy, and exhibit various personal interests. To address this challenge, we propose a novel unified topic-guided encoder-decoder (UTGED) framework; it models latent topics to reflect salient user interest, whose topic mixture then guides encoding a user's history and topic words control decoding their self-introduction. For experiments, we collect a large-scale Twitter dataset, and extensive results show the superiority of our UTGED to the advanced encoder-decoder models without topic modeling.

cs.CL↗
arXiv2023

Global warming in the pipeline

Improved knowledge of glacial-to-interglacial global temperature change implies that fast-feedback equilibrium climate sensitivity (ECS) is 1.2 +/- 0.3°C (2$σ$) per W/m$^2$. Consistent analysis of temperature over the full Cenozoic era -- including "slow" feedbacks by ice sheets and trace gases -- supports this ECS and implies that CO$_2$ was about 300 ppm in the Pliocene and 400 ppm at transition to a nearly ice-free planet, thus exposing unrealistic lethargy of ice sheet models. Equilibrium global warming including slow feedbacks for today's human-made greenhouse gas (GHG) climate forcing (4.1 W/m$^2$) is 10°C, reduced to 8°C by today's aerosols. Decline of aerosol emissions since 2010 should increase the 1970-2010 global warming rate of 0.18°C per decade to a post-2010 rate of at least 0.27°C per decade. Under the current geopolitical approach to GHG emissions, global warming will likely pierce the 1.5°C ceiling in the 2020s and 2°C before 2050. Impacts on people and nature will accelerate as global warming pumps up hydrologic extremes. The enormity of consequences demands a return to Holocene-level global temperature. Required actions include: 1) a global increasing price on GHG emissions, 2) East-West cooperation in a way that accommodates developing world needs, and 3) intervention with Earth's radiation imbalance to phase down today's massive human-made "geo-transformation" of Earth's climate. These changes will not happen with the current geopolitical approach, but current political crises present an opportunity for reset, especially if young people can grasp their situation.

physics.ao-ph↗
arXiv2023

M$^{6}$Doc: A Large-Scale Multi-Format, Multi-Type, Multi-Layout, Multi-Language, Multi-Annotation Category Dataset for Modern Document Layout Analysis

Document layout analysis is a crucial prerequisite for document understanding, including document retrieval and conversion. Most public datasets currently contain only PDF documents and lack realistic documents. Models trained on these datasets may not generalize well to real-world scenarios. Therefore, this paper introduces a large and diverse document layout analysis dataset called $M^{6}Doc$. The $M^6$ designation represents six properties: (1) Multi-Format (including scanned, photographed, and PDF documents); (2) Multi-Type (such as scientific articles, textbooks, books, test papers, magazines, newspapers, and notes); (3) Multi-Layout (rectangular, Manhattan, non-Manhattan, and multi-column Manhattan); (4) Multi-Language (Chinese and English); (5) Multi-Annotation Category (74 types of annotation labels with 237,116 annotation instances in 9,080 manually annotated pages); and (6) Modern documents. Additionally, we propose a transformer-based document layout analysis method called TransDLANet, which leverages an adaptive element matching mechanism that enables query embedding to better match ground truth to improve recall, and constructs a segmentation branch for more precise document image instance segmentation. We conduct a comprehensive evaluation of $M^{6}Doc$ with various layout analysis methods and demonstrate its effectiveness. TransDLANet achieves state-of-the-art performance on $M^{6}Doc$ with 64.5% mAP. The $M^{6}Doc$ dataset will be available at https://github.com/HCIILAB/M6Doc.

cs.CV↗
arXiv2023

Disintegration of Long-Period Comet C/2021 A1 (Leonard)

We present imaging observations of the disintegrating long-period comet C/2021 A1 (Leonard). High resolution observations with Hubble Space Telescope show no evidence for surviving fragments, and place a 3 sigma upper limit to their possible radius about 60 m (albedo 0.1 assumed). In contrast, wide field observations from the Swan Hill Observatory, Australia, show an extensive debris cloud, the cross-section and estimated mass of which are consistent with complete disintegration of the nucleus near mid- December 2021 (at about 0.8 au). Two methods give the pre-disruption nucleus radius, r = 0.6+/-0.2 km. Tidal, collisional, sublimation and pressure-confined explosion models provide implausible explanations of the disintegration. However, rotational instability driven by outgassing torques has a very short timescale (of order 0.1 year) given the orbit and size of the C/2021 A1 nucleus, and offers the most plausible mechanism for the disruption. Initial rotational breakup is accelerated by the exposure and strong sublimation of previously buried volatiles, leading to catastrophic destruction of the nucleus.

astro-ph.EP↗
arXiv2023

Atomically-precise Vacancy-assembled Quantum Antidots

Patterning antidots ("voids") into well-defined antidot lattices creates an intriguing class of artificial structures for the periodic modulation of 2D electron systems, leading to anomalous transport properties and exotic quantum phenomena as well as enabling the precise bandgap engineering of 2D materials to address technological bottleneck issues. However, realizing such atomic-scale quantum antidots (QADs) is infeasible by current nanolithographic techniques. Here, we report an atomically-precise bottom-up fabrication of a series of atomic-scale QADs with elegantly engineered quantum states through a controllable assembly of a chalcogenide single vacancy (SV) in 2D PtTe2, a type-II Dirac semimetal. Te SVs as atomic-scale "antidots" undergo thermal migration and assembly into highly-ordered SV lattices spaced by a single Te atom, reaching the ultimate downscaling limit of antidot lattices. Increasing the number of SVs in QADs strengthens the cumulative repulsive potential and consequently enhances collective interference of multiple-pocket scattered quasiparticles inside QADs, creating multi-level quantum hole states with tunable gap from telecom to far-infrared regime. Moreover, precisely engineered quantum hole states of QADs are symmetry-protected and thus survive upon atom-by-atom oxygen substitutional doping. Therefore, SV-assembled QADs exhibit unprecedented robustness and property tunability, which not only holds the key to their future applications but also embody a wide variety of material technologies.

cond-mat.mes-hall↗