arXiv subjects

Yi Lu

Publications and source records attributed to Yi Lu.

At least 163 records · Page 9Linked to original sources

arXiv2023

Visualization of Oxygen Vacancies and Self-doped Ligand Holes in La3Ni2O7-δ

The recent discovery of superconductivity in La3Ni2O7-δ under high pressure with a transition temperature around 80 K has sparked extensive experimental and theoretical efforts. Several key questions regarding the pairing mechanism remain to be answered, such as the most relevant atomic orbitals and the role of atomic deficiencies. Here, we develop a new energy-filtered multislice electron ptychography technique, assisted with electron energy loss spectroscopy, to address these critical issues. Oxygen vacancies are directly visualized and are found to primarily occupy the inner apical sites, which have been proposed to be crucial to superconductivity. We precisely determine the nanoscale stoichiometry and its correlation to the oxygen K edge spectra, which reveals a significant inhomogeneity in the oxygen content and electronic structure within the sample. The spectroscopic results also unveil that stoichiometric La3Ni2O7 is in the charge-transfer regime, with holes that are self-doped from Ni sites into O sites. The outer apical oxygen is found to be less relevant to the low-energy physics and can be safely disregarded in theoretical models. These observations will assist in further development and understanding of superconducting nickelate materials. Our imaging technique for quantifying atomic deficiencies can also be widely applied in materials science and condensed matter physics.

cond-mat.supr-con↗

arXiv2023

Making Harmful Behaviors Unlearnable for Large Language Models

Large language models (LLMs) have shown great potential as general-purpose AI assistants in various domains. To meet the requirements of different applications, LLMs are often customized by further fine-tuning. However, the powerful learning ability of LLMs not only enables them to acquire new tasks but also makes them susceptible to learning undesired behaviors. For example, even safety-aligned LLMs can be easily fine-tuned into harmful assistants as the fine-tuning data often contains implicit or explicit harmful content. Can we train LLMs on harmful data without learning harmful behaviors? This paper proposes a controllable training framework that makes harmful behaviors unlearnable during the fine-tuning process. Specifically, we introduce ``security vectors'', a few new parameters that can be separated from the LLM, to ensure LLM's responses are consistent with the harmful behavior. Security vectors are activated during fine-tuning, the consistent behavior makes LLM believe that such behavior has already been learned, there is no need to further optimize for harmful data. During inference, we can deactivate security vectors to restore the LLM's normal behavior. The experimental results show that the security vectors generated by 100 harmful samples are enough to prevent LLM from learning 1000 harmful samples, while preserving the ability to learn other useful information.

cs.LG↗

arXiv2023

QCD equation of state and thermodynamic observables from computationally minimal Dyson-Schwinger Equations

We study the QCD equation of state and other thermodynamic observables including the isentropic trajectories and the speed of sound. These observables are of eminent importance for the understanding of experimental results in heavy ion collisions and also provide a QCD input for studies of the timeline of heavy-ion-collisions with hydrodynamical simulations. They can be derived from the quark propagator whose gap equation is solved within a minimal approximation to the Dyson-Schwinger equations of QCD at finite temperature and density. This minimal approximation aims at a combination of computational efficiency and simplification of the truncation scheme while maintaining quantitative precision. This minimal DSE scheme is confronted and benchmarked with results for correlation functions and observables from first principles QCD lattice at vanishing density and quantitative functional approaches at finite density.

hep-ph↗

arXiv2023

Constructing the Equation of State of QCD in a functional QCD based scheme

We construct the equation of state (EoS) of QCD based on the finite chemical potential information from the functional QCD approaches, with the assistance of the lattice QCD EoS. The obtained EoS is consistent with the up-to-date estimations of the QCD phase diagram, including a phase transition temperature at zero chemical potential of $T=155$ MeV, the curvature of the transition line $κ=0.016$ and also a critical end point at $(T,μ_B)=(118, 600)$ MeV. In specific, the phase diagram mapping is achieved by incorporating the order parameters into the EoS, namely the dynamical quark mass for the chiral phase transition together with the Polyakov loop parameter for the deconfinement phase transition. We also implement the EoS in hydrodynamic simulations to compute the particle yields, ratios and collective flow, and find that our obtained EoS agrees well with the commonly used one based on the combination of lattice QCD simulation and hadron resonance gas model.

hep-ph↗

arXiv2023

Can 5G NR Sidelink communications support wireless augmented reality?

Smart glasses that support augmented reality (AR) have the potential to become the consumer's primary medium of connecting to the future internet. For the best quality of user experience, AR glasses must have a small form factor and long battery life, while satisfying the data rate and latency requirements of AR applications. To extend the AR glasses' battery life, the computation and processing involved in AR may be offloaded to a companion device, such as a smartphone, through a wireless connection. Sidelink (SL), i.e., the D2D communication interface of 5G NR, is a potential candidate for this wireless link. In this paper, we use system-level simulations to analyze the feasibility of NR SL for supporting AR. Our simulator incorporates the PHY layer structure and MAC layer resource scheduling of 3GPP SL, standard 3GPP channel models, and MCS configurations. Our results suggest that the current SL standard specifications are insufficient for high-end AR use cases with heavy interaction but can support simpler previews and file transfers. We further propose two enhancements to SL resource allocation, which have the potential to offer significant performance improvements for AR applications.

cs.NI↗

arXiv2023

Sphaleron freeze-in baryogenesis with gravitational waves from the QCD transition

A large primordial lepton asymmetry is capable of explaining the baryon asymmetry of the Universe (BAU) through suppression of the electroweak sphaleron rates (``sphaleron freeze-in") which can lead to a first-order cosmic QCD transition with an observable gravitational wave (GW) signal. With next-to-leading order dimensional reduction and the exact 1-loop fluctuation determinant, we accurately compute the lepton asymmetry needed to realize this paradigm, finding it to be an order of magnitude smaller than previous estimates. Further, we apply an improved QCD equation of state capable of describing the phase transition line together with the critical endpoint leading to better agreement with lattice and functional QCD results. Based on this, we identify the range of lepton flavor asymmetries inducing a first-order cosmic QCD transition. We then extract the parameters relevant to the prediction of GW signal from a first-order cosmic QCD transition. Our result showcases the possibility of probing the sphaleron freeze-in paradigm as an explanation of BAU by future gravitational wave experiments like $μ$Ares.

hep-ph↗

arXiv2023

Nonparametric Decoding for Generative Retrieval

The generative retrieval model depends solely on the information encoded in its model parameters without external memory, its information capacity is limited and fixed. To overcome the limitation, we propose Nonparametric Decoding (Np Decoding) which can be applied to existing generative retrieval models. Np Decoding uses nonparametric contextualized vocab embeddings (external memory) rather than vanilla vocab embeddings as decoder vocab embeddings. By leveraging the contextualized vocab embeddings, the generative retrieval model is able to utilize both the parametric and nonparametric space. Evaluation over 9 datasets (8 single-hop and 1 multi-hop) in the document retrieval task shows that applying Np Decoding to generative retrieval models significantly improves the performance. We also show that Np Decoding is data- and parameter-efficient, and shows high performance in the zero-shot setting.

cs.IR↗

arXiv2023

Absence of localized $5d^1$ electrons in KTaO$_3$ interface superconductors

Recently, an exciting discovery of orientation-dependent superconductivity was made in two-dimensional electron gas (2DEG) at the interfaces of LaAlO$_3$/KTaO$_3$ (LAO/KTO) or EuO/KTaO$_3$ (EuO/KTO). The superconducting transition temperature can reach a $T_c$ of up to $\sim$ 2.2 K, which is significantly higher than its 3$d$ counterpart LaAlO$_3$/SrTiO$_3$ (LAO/STO) with a $T_c$ of $\sim$ 0.2 K. However, the underlying origin remains to be understood. To uncover the nature of electrons in KTO-based interfaces, we employ x-ray absorption spectroscopy (XAS) and resonant inelastic x-ray spectroscopy (RIXS) to study LAO/KTO and EuO/KTO with different orientations. We reveal the absence of $dd$ orbital excitations in all the measured samples. Our RIXS results are well reproduced by calculations that considered itinerant $5d$ electrons hybridized with O $2p$ electrons. This suggests that there is a lack of localized Ta $5d^1$ electrons in KTO interface superconductors, which is consistent with the absence of magnetic hysteresis observed in magneto-resistance (MR) measurements. These findings offer new insights into our understanding of superconductivity in Ta $5d$ interface superconductors and their potential applications.

cond-mat.supr-con↗

arXiv2023

The Short-term Impact of Congestion Taxes on Ridesourcing Demand and Traffic Congestion: Evidence from Chicago

Ridesourcing is popular in many cities. Despite its theoretical benefits, a large body of studies have claimed that ridesourcing also brings (negative) externalities (e.g., inducing trips and aggravating traffic congestion). Therefore, many cities are planning to enact or have already enacted policies to regulate its use. However, these policies' effectiveness or impact on ridesourcing demand and traffic congestion is uncertain. To this end, this study applies difference-in-differences (i.e., a regression-based causal inference approach) to empirically evaluate the effects of the congestion tax policy on ridesourcing demand and traffic congestion in Chicago. It shows that this congestion tax policy significantly curtails overall ridesourcing demand but marginally alleviates traffic congestion. The results are robust to the choice of time windows and data sets, additional control variables, alternative model specifications, alternative control groups, and alternative modeling approaches (i.e., regression discontinuity in time). Moreover, considerable heterogeneity exists. For example, the policy notably reduces ridesourcing demand with short travel distances, but such an impact is gradually attenuated as the distance increases.

econ.GN↗

arXiv2023

Improved Knowledge Distillation for Pre-trained Language Models via Knowledge Selection

Knowledge distillation addresses the problem of transferring knowledge from a teacher model to a student model. In this process, we typically have multiple types of knowledge extracted from the teacher model. The problem is to make full use of them to train the student model. Our preliminary study shows that: (1) not all of the knowledge is necessary for learning a good student model, and (2) knowledge distillation can benefit from certain knowledge at different training steps. In response to these, we propose an actor-critic approach to selecting appropriate knowledge to transfer during the process of knowledge distillation. In addition, we offer a refinement of the training algorithm to ease the computational burden. Experimental results on the GLUE datasets show that our method outperforms several strong knowledge distillation baselines significantly.

cs.CL↗

arXiv2023

Evolution of orbital excitations from insulating to superconducting MgTi$_2$O$_4$ films

Spinel oxides are well-known functional materials but rarely show superconductivity. Recently, emergent superconductivity was discovered in MgTi$_2$O$_4$, which is attributed to the increase of electron doping and the suppression of orbital order. Here, we utilized Ti $L$-edge resonant inelastic X-ray scattering to study the orbital excitations in superconducting (SC) and insulating MgTi$_2$O$_4$ films. We find that the spectral weight of orbital excitations is enhanced and the energy of $t_{2g}$ intra-band excitation is softened in the SC film compared to the insulating one, suggesting higher electron doping and suppressed orbital order gap in the SC sample. These observations were further supported by our multiplet calculations using minimal two-site model. Our results provide spectroscopic evidence for the competition between orbital order and superconductivity in MgTi$_2$O$_4$ and shed light on searching for novel superconductors in spinel oxides.

cond-mat.supr-con↗

arXiv2023

Side-by-Side vs Face-to-Face: Evaluating Colocated Collaboration via a Transparent Wall-sized Display

Traditional wall-sized displays mostly only support side-by-side co-located collaboration, while transparent displays naturally support face-to-face interaction. Many previous works assume transparent displays support collaboration. Yet it is unknown how exactly its afforded face-to-face interaction can support loose or close collaboration, especially compared to the side-by-side configuration offered by traditional large displays. In this paper, we used an established experimental task that operationalizes different collaboration coupling and layout locality, to compare pairs of participants collaborating side-by-side versus face-to-face in each collaborative situation. We compared quantitative measures and collected interview and observation data to further illustrate and explain our observed user behavior patterns. The results showed that the unique face-to-face collaboration brought by transparent display can result in more efficient task performance, different territorial behavior, and both positive and negative collaborative factors. Our findings provided empirical understanding about the collaborative experience supported by wall-sized transparent displays and shed light on its future design.

cs.HC↗

arXiv2022

COGEDAP: A COmprehensive GEnomic Data Analysis Platform

Non-sharable sensitive data collection and analysis in large-scale consortia for genomic research is complicated. Time consuming issues in installing software arise due to different operating systems, software dependencies and running the software. Therefore, easier, more standardized, automated protocols and platforms can be a solution to overcome these issues. We have developed one such solution for genomic data analysis using software container technologies. The platform, COGEDAP, consists of different software tools placed into Singularity containers with corresponding pipelines and instructions on how to perform genome-wide association studies (GWAS) and other genomic data analysis via corresponding tools. Using a provided helper script written in Python, users can obtain auto-generated scripts to conduct the desired analysis both on high-performance computing (HPC) systems and on personal computers. The analyses can be done by running these auto-generated scripts with the software containers. The helper script also performs minor re-formatting of the input/output data, so that the end user can work with a unified file format regardless of which genetic software is used for the analysis. COGEDAP is actively being used by users from different countries/projects to conduct their genomic data analyses. Thanks to this platform, users can easily run GWAS and other genomic analyses without spending much effort on software installation, data formats, and other technical requirements.

q-bio.GN↗

arXiv2022

Joint RIS Calibration and Multi-User Positioning

Reconfigurable intelligent surfaces (RISs) are expected to be a key component enabling the mobile network evolution towards a flexible and intelligent 6G wireless platform. In most of the research works so far, RIS has been treated as a passive base station (BS) with a known state, in terms of its location and orientation, to boost the communication and/or terminal positioning performance. However, such performance gains cannot be guaranteed anymore when the RIS state is not perfectly known. In this paper, by taking the RIS state uncertainty into account, we formulate and study the performance of a joint RIS calibration and user positioning (JrCUP) scheme. From the Fisher information perspective, we formulate the JrCUP problem in a network-centric single-input multiple-output (SIMO) scenario with a single BS, and derive the analytical lower bound for the states of both user and RIS. We also demonstrate the geometric impact of different user locations on the JrCUP performance while also characterizing the performance under different RIS sizes. Finally, the study is extended to a multi-user scenario, shown to further improve the state estimation performance.

eess.SP↗

arXiv2022

Large-scale Atomistic Simulation of Quantum Effects in SrTiO$_3$ from First Principles

Quantum effects of lattice vibration play a major role in many physical properties of condensed matter systems, including thermal properties such as specific heat, structural phase transition, as well as phenomena such as quantum crystal and quantum paraelectricity that are closely related to zero-point fluctuations. However, realizing atomistic simulations for realistic materials with a fully quantum-mechanical description remains a great challenge. Here, we propose a first-principle strategy for large scale Molecular Dynamics simulation, where high accuracy force field obtained by Deep-Potential (DP) is combined with Quantum Thermal Bath (QTB) method to account for quantum effects. We demonstrate the power of this DP+QTB method using the archetypal example SrTiO$_3$, which exhibits several phenomena induced by quantum fluctuations, such as the suppressed structure phase transition temperature, the quantum paraelectric ground state at low temperature and the quantum critical behavior $1/T^2$ law of dielectric constant. Our DP+QTB strategy is efficient in simulating large scale system, and is first principle. More importantly, quantum effects of other systems could also be investigated as long as corresponding DP model is trained. This strategy would greatly enrich our vision and means to study quantum behavior of condensed matter physics.

cond-mat.mtrl-sci↗

arXiv2022

Revealing the Signal of QCD Phase Transition in Heavy-Ion Collisions

We propose a novel method to construct the Landau thermodynamic potential directly from the fluctuations measured in heavy-ion collisions. The potential is capable of revealing the signal of the critical end-point (CEP) and the first order phase transition (FOPT) of QCD in the system even away from the phase transition region. With the available experimental data, we show that the criterion of the FOPT is negative for most of the collision energies which indicates no signal of FOPT. The data at $\sqrt{s_\mathrm{NN}}=7.7$ GeV with 0-5% centrality shows a different behavior and the mean value of the data satisfies the criterion. However, the uncertainty is still too large to make a certain conclusion. The higher order fluctuations are also required for confirming the signal. We emphasize therefore that new measurements with higher precision for the $C_{1,...,6}$ within 0-5% centrality in the vicinity of $\sqrt{s_\mathrm{NN}}=7.7$ GeV are in demand which may finally reveal the signal of QCD phase transition.

hep-ph↗

arXiv2022

High spectral-resolution interferometry down to 1 micron with Asgard/BIFROST at VLTI: Science drivers and project overview

We present science cases and instrument design considerations for the BIFROST instrument that will open the short-wavelength (Y/J/H-band), high spectral dispersion (up to R=25,000) window for the VLT Interferometer. BIFROST will be part of the Asgard Suite of instruments and unlock powerful venues for studying accretion & mass-loss processes at the early/late stages of stellar evolution, for detecting accreting protoplanets around young stars, and for probing the spin-orbit alignment in directly-imaged planetary systems and multiple star systems. Our survey on GAIA binaries aims to provide masses and precision ages for a thousand stars, providing a legacy data set for improving stellar evolutionary models as well as for Galactic Archaeology. BIFROST will enable off-axis spectroscopy of exoplanets in the 0.025-1" separation range, enabling high-SNR, high spectral resolution follow-up of exoplanets detected with ELT and JWST. We give an update on the status of the project, outline our key technology choices, and discuss synergies with other instruments in the proposed Asgard Suite of instruments.

astro-ph.IM↗

arXiv2022

Origin of negative thermal expansion and pressure induced amorphization in zirconium tungstate from machine-learning potential

Understanding various macroscopic pressure-volume-temperature properties of materials on the atomistic level has always been an ambition for physicists and material scientists. Particularly, some materials such as zirconium tungstate (ZrW2O8), exhibit multiple exotic properties including negative thermal expansion (NTE) and pressure-induced amorphization (PIA). Here, using machine-learning based deep potential, we trace both of the phenomena in ZrW2O8 back to a common atomistic origin, where the nonbridging O atoms play a critical role. We demonstrate that the nonbridging O atoms confer great flexibility to vibration of polyhedrons, and kinetically drive volume shrinking on heating, or NTE. In addition, beyond a certain critical pressure, we find that the migration of nonbridging O atoms leads to additional bond formation that lowers the potential energy, suggesting that the PIA is a potential-driven first-order phase transition. Most importantly, we identify a second critical pressure beyond which the amorphous phase of ZrW2O8 undergoes a hidden phase transition from a reversible phase to an irreversible one.

cond-mat.mtrl-sci↗