Bing Zhu
Research Subjects
Research Interests
Bing Zhu has a total of 97 co-authors. Here are the top co-authors by number of shared publications:
- Matthias Weidemüller
Universität Heidelberg
14 shared publications - Yuhai Jiang
ShanghaiTech University
9 shared publications Canzhu Tan
CAS Center For Excellence in Quantum Information and Quantum Physics, University of Science and Technology of China, Hefei 230026, China
8 shared publications- Dajun Wang
Chinese University of Hong Kong
6 shared publications Fachao Hu
Hefei National Laboratory for Physical Sciences at the Microscale University of Science and Technology of China 230026 Hefei Auhui China
5 shared publicationsFudong Wang
Department of Physics, The Chinese University of Hong Kong, Hong Kong, China
5 shared publications
Latest Publications
- Open Access
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 5322-5333
ClimateIQA: A New Dataset and Benchmark to Advance Vision-Language Models in Meteorology Anomalies Analysis
- 3 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 5322-5333
Open AccessClimateIQA: A New Dataset and Benchmark to Advance Vision-Language Models in Meteorology Anomalies Analysis
- 23 September 2024
- Published by American Physical Society (APS) in Physical Review A
- Vol. 110 (3) , 033114
Multichannel photoionization of cold strontium atoms
The multichannel single photoionization of cold strontium atoms is investigated experimentally and theoretically with 800-nm femtosecond laser pulses of 35-fs duration at laser intensities of 3–6 . The photoelectron energy spectra exhibit rich multipeak structures, which are identified as the multichannel ionization of the ground state and excited states , , and . The photoelectron momentum distributions reveal distinct structures, and the corresponding photoelectron angular distributions provide valuable insights into the multiphoton ionization dynamics. As the laser intensity increases, the locations of the photoelectron energy peaks originating from the ionization of the excited states and remain unshifted, indicating the existence of Freeman resonance. The profiles from numerical solutions of the time-dependent Schrödinger equation (TDSE) are generally in agreement with the experimental results. However, some discrepancies, i.e., the ratio of partial waves contributing to the final states and tiny structures caused by resonance, may arise from the incomplete modeling of the valence-electron correlation effect in the present TDSE simulations based on the single-active-electron approximation.- 23 September 2024
- Published by American Physical Society (APS) in Physical Review A
- Vol. 110 (3) , 033114
Multichannel photoionization of cold strontium atoms
The multichannel single photoionization of cold strontium atoms is investigated experimentally and theoretically with 800-nm femtosecond laser pulses of 35-fs duration at laser intensities of 3–6 . The photoelectron energy spectra exhibit rich multipeak structures, which are identified as the multichannel ionization of the ground state and excited states , , and . The photoelectron momentum distributions reveal distinct structures, and the corresponding photoelectron angular distributions provide valuable insights into the multiphoton ionization dynamics. As the laser intensity increases, the locations of the photoelectron energy peaks originating from the ionization of the excited states and remain unshifted, indicating the existence of Freeman resonance. The profiles from numerical solutions of the time-dependent Schrödinger equation (TDSE) are generally in agreement with the experimental results. However, some discrepancies, i.e., the ratio of partial waves contributing to the final states and tiny structures caused by resonance, may arise from the incomplete modeling of the valence-electron correlation effect in the present TDSE simulations based on the single-active-electron approximation.- Open Access
FinTextQA: A Dataset for Long-form Financial Question Answering
Accurate evaluation of financial question answering (QA) systems necessitates a comprehensive dataset encompassing diverse question types and contexts. However, current financial QA datasets lack scope diversity and question complexity. This work introduces FinTextQA, a novel dataset for long-form question answering (LFQA) in finance. FinTextQA comprises 1,262 high-quality, source-attributed QA pairs extracted and selected from finance textbooks and government agency websites.Moreover, we developed a Retrieval-Augmented Generation (RAG)-based LFQA system, comprising an embedder, retriever, reranker, and generator. A multi-faceted evaluation approach, including human ranking, automatic metrics, and GPT-4 scoring, was employed to benchmark the performance of different LFQA system configurations under heightened noisy conditions. The results indicate that: (1) Among all compared generators, Baichuan2-7B competes closely with GPT-3.5-turbo in accuracy score; (2) The most effective system configuration on our dataset involved setting the embedder, retriever, reranker, and generator as Ada2, Automated Merged Retrieval, Bge-Reranker-Base, and Baichuan2-7B, respectively; (3) models are less susceptible to noise after the length of contexts reaching a specific threshold.
Open AccessFinTextQA: A Dataset for Long-form Financial Question Answering
Accurate evaluation of financial question answering (QA) systems necessitates a comprehensive dataset encompassing diverse question types and contexts. However, current financial QA datasets lack scope diversity and question complexity. This work introduces FinTextQA, a novel dataset for long-form question answering (LFQA) in finance. FinTextQA comprises 1,262 high-quality, source-attributed QA pairs extracted and selected from finance textbooks and government agency websites.Moreover, we developed a Retrieval-Augmented Generation (RAG)-based LFQA system, comprising an embedder, retriever, reranker, and generator. A multi-faceted evaluation approach, including human ranking, automatic metrics, and GPT-4 scoring, was employed to benchmark the performance of different LFQA system configurations under heightened noisy conditions. The results indicate that: (1) Among all compared generators, Baichuan2-7B competes closely with GPT-3.5-turbo in accuracy score; (2) The most effective system configuration on our dataset involved setting the embedder, retriever, reranker, and generator as Ada2, Automated Merged Retrieval, Bge-Reranker-Base, and Baichuan2-7B, respectively; (3) models are less susceptible to noise after the length of contexts reaching a specific threshold.
- 20 February 2024
- Published by American Physical Society (APS) in Physical Review A
- Vol. 109 (2) , 023118
Magneto-optical trap reaction microscope for photoionization of cold strontium atoms
We developed a magneto-optical trap reaction microscope (MOTREMI) for strontium atoms by combining multiparticle coincident detection with the laser cooling technique. The present compact injection system enables the production of cold Sr atoms in three modes of two-dimensional (2D) magneto-optical trap (MOT) beam, molasses, and three-dimensional (3D) MOT, providing adjustable densities and ratios of the ground state () and the excited states and . The target profiles including temperature, density, and size of the 3D MOT as well as the 2D MOT beam cold atomic flux were characterized in detail. Using the present setup, we investigate the single photoionization of Sr atoms with molasses through few-photons absorption in 800 nm, where and are detected in coincidence and most of the ionization channels are identified. The momentum resolution for coincident and along time-of-flight direction are achieved up to 0.12 a.u. and 0.02 a.u., respectively. The spectra of photoelectron ionized from the ground state as well as several excited states show distinct structures manifesting prominent contributions from multiphoton absorption. Detecting complete vector momenta of electrons and recoil ions in coincidence paves a way for further investigation of two-electron correlation dynamics and multielectron effects with alkaline-earth atoms.- 20 February 2024
- Published by American Physical Society (APS) in Physical Review A
- Vol. 109 (2) , 023118
Magneto-optical trap reaction microscope for photoionization of cold strontium atoms
We developed a magneto-optical trap reaction microscope (MOTREMI) for strontium atoms by combining multiparticle coincident detection with the laser cooling technique. The present compact injection system enables the production of cold Sr atoms in three modes of two-dimensional (2D) magneto-optical trap (MOT) beam, molasses, and three-dimensional (3D) MOT, providing adjustable densities and ratios of the ground state () and the excited states and . The target profiles including temperature, density, and size of the 3D MOT as well as the 2D MOT beam cold atomic flux were characterized in detail. Using the present setup, we investigate the single photoionization of Sr atoms with molasses through few-photons absorption in 800 nm, where and are detected in coincidence and most of the ionization channels are identified. The momentum resolution for coincident and along time-of-flight direction are achieved up to 0.12 a.u. and 0.02 a.u., respectively. The spectra of photoelectron ionized from the ground state as well as several excited states show distinct structures manifesting prominent contributions from multiphoton absorption. Detecting complete vector momenta of electrons and recoil ions in coincidence paves a way for further investigation of two-electron correlation dynamics and multielectron effects with alkaline-earth atoms.- Open Access
Software Code Quality Measurement: Implications from Metric Distributions
Software code quality is a construct with three dimensions: maintainability, reliability, and functionality. Although many firms have incorporated code quality metrics in their operations, evaluating these metrics still lacks consistent standards. We categorized distinct metrics into two types: 1) monotonic metrics that consistently influence code quality; and 2) non-monotonic metrics that lack a consistent relationship with code quality. To consistently evaluate them, we proposed a distribution-based method to get metric scores. Our empirical analysis includes 36,460 high-quality open-source software (OSS) repositories and their raw metrics from SonarQube and CK. The evaluated scores demonstrate great explainability on software adoption. Our work contributes to the multi-dimensional construct of code quality and its metric measurements, which provides practical implications for consistent measurements on both monotonic and non-monotonic metrics.
Open AccessSoftware Code Quality Measurement: Implications from Metric Distributions
Software code quality is a construct with three dimensions: maintainability, reliability, and functionality. Although many firms have incorporated code quality metrics in their operations, evaluating these metrics still lacks consistent standards. We categorized distinct metrics into two types: 1) monotonic metrics that consistently influence code quality; and 2) non-monotonic metrics that lack a consistent relationship with code quality. To consistently evaluate them, we proposed a distribution-based method to get metric scores. Our empirical analysis includes 36,460 high-quality open-source software (OSS) repositories and their raw metrics from SonarQube and CK. The evaluated scores demonstrate great explainability on software adoption. Our work contributes to the multi-dimensional construct of code quality and its metric measurements, which provides practical implications for consistent measurements on both monotonic and non-monotonic metrics.
- Open Access
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 6025-6047
FinTextQA: A Dataset for Long-form Financial Question Answering
Jian Chen, Peilin Zhou, Yining Hua, Loh Xin, Kehui Chen, Ziyuan Li, Bing Zhu, Junwei Liang. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 6025-6047
Open AccessFinTextQA: A Dataset for Long-form Financial Question Answering
Jian Chen, Peilin Zhou, Yining Hua, Loh Xin, Kehui Chen, Ziyuan Li, Bing Zhu, Junwei Liang. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.
- Open Access
- 1 January 2024
- Published by Springer Nature in Financial Innovation
- Vol. 10 (1) , 1-30
Relationships among return and liquidity of cryptocurrencies
The cryptocurrency market is a complex and rapidly evolving financial landscape in which understanding the inter- and intra-asset dependencies among key financial variables, such as return and liquidity, is crucial. In this study, we analyze daily return and liquidity data for six major cryptocurrencies, namely Bitcoin, Ethereum, Ripple, Binance Coin, Litecoin, and Dogecoin, spanning the period from June 3, 2020, to November 30, 2022. Liquidity is estimated using three low-frequency proxies: the Amihud ratio and the Abdi and Ranaldo (AR) and Corwin and Schultz (CS) estimators. To account for autoregressive and persistent effects, we apply the autoregressive integrated moving average-generalized autoregressive conditional heteroscedasticity (ARIMA-GARCH) model and subsequently utilize the copula method to examine the interdependent relationships between the return on and liquidity of the six cryptocurrencies. Our analysis reveals strong cross-asset lower-tail dependence in return and significant cross-asset upper-tail dependence in illiquidity measures, with more pronounced dependence observed in specific cryptocurrency pairs, primarily involving Bitcoin, Ethereum, and Litecoin. We also observe that returns tend to be higher when liquidity is lower in the cryptocurrency market. Our findings have significant implications for portfolio diversification, asset allocation, risk management, and trading strategy development for investors and traders, as well as regulatory policy-making for regulators. This study contributes to a deeper understanding of the cryptocurrency marketplace and can help inform investment decision making and regulatory policies in this emerging financial domain.
- 1 January 2024
- Published by Springer Nature in Financial Innovation
- Vol. 10 (1) , 1-30
Open AccessRelationships among return and liquidity of cryptocurrencies
The cryptocurrency market is a complex and rapidly evolving financial landscape in which understanding the inter- and intra-asset dependencies among key financial variables, such as return and liquidity, is crucial. In this study, we analyze daily return and liquidity data for six major cryptocurrencies, namely Bitcoin, Ethereum, Ripple, Binance Coin, Litecoin, and Dogecoin, spanning the period from June 3, 2020, to November 30, 2022. Liquidity is estimated using three low-frequency proxies: the Amihud ratio and the Abdi and Ranaldo (AR) and Corwin and Schultz (CS) estimators. To account for autoregressive and persistent effects, we apply the autoregressive integrated moving average-generalized autoregressive conditional heteroscedasticity (ARIMA-GARCH) model and subsequently utilize the copula method to examine the interdependent relationships between the return on and liquidity of the six cryptocurrencies. Our analysis reveals strong cross-asset lower-tail dependence in return and significant cross-asset upper-tail dependence in illiquidity measures, with more pronounced dependence observed in specific cryptocurrency pairs, primarily involving Bitcoin, Ethereum, and Litecoin. We also observe that returns tend to be higher when liquidity is lower in the cryptocurrency market. Our findings have significant implications for portfolio diversification, asset allocation, risk management, and trading strategy development for investors and traders, as well as regulatory policy-making for regulators. This study contributes to a deeper understanding of the cryptocurrency marketplace and can help inform investment decision making and regulatory policies in this emerging financial domain.
- Open Access
Magneto-optical trap reaction microscope for photoionization of cold strontium atoms
We developed a magneto-optical trap reaction microscope (MOTREMI) for strontium atoms by combining the multi-particle coincident detection with laser cooling technique. Present compact injection system can provide cold Sr atoms in three modes of 2D MOT, molasses and 3D MOT, delivering targets with adjustable densities and ratios of the ground state $5s^2$ ($^1S_{0}$) and the excited states $5s5p$ ($^{1}P_{1}$ and $^{3}P_{J}$ etc). The target profiles for the temperature, the density and the size of 3D MOT as well as cold atomic flux in 2D MOT model were characterized in details. With present state-of-the-art setup, we demonstrated the single photoionization of Sr atoms with molasses by absorption of few 800-nm photons, where Sr$^+$ and $e$ were detected in coincidence and most of ionization channels were identified taking into account photoelectron energy, laser-intensity dependence, and target dependence. The best momentum resolution of coincident Sr$^+$ and $e$ along time-of-flight are achieved up to 0.12 a.u. and 0.02 a.u., respectively. Present photoelectron momentum distributions ionized from the ground state and a few excited states illuminate unprecedentedly rich landscapes manifesting prominent features for multi-photon absorption. The full vector momenta of electrons and recoil ion in coincidence paves the way to further studying two-electron correlation dynamics and multi-electron effects in the multiple ionization of alkaline-earth atoms in the ultraviolet region.Open AccessMagneto-optical trap reaction microscope for photoionization of cold strontium atoms
We developed a magneto-optical trap reaction microscope (MOTREMI) for strontium atoms by combining the multi-particle coincident detection with laser cooling technique. Present compact injection system can provide cold Sr atoms in three modes of 2D MOT, molasses and 3D MOT, delivering targets with adjustable densities and ratios of the ground state $5s^2$ ($^1S_{0}$) and the excited states $5s5p$ ($^{1}P_{1}$ and $^{3}P_{J}$ etc). The target profiles for the temperature, the density and the size of 3D MOT as well as cold atomic flux in 2D MOT model were characterized in details. With present state-of-the-art setup, we demonstrated the single photoionization of Sr atoms with molasses by absorption of few 800-nm photons, where Sr$^+$ and $e$ were detected in coincidence and most of ionization channels were identified taking into account photoelectron energy, laser-intensity dependence, and target dependence. The best momentum resolution of coincident Sr$^+$ and $e$ along time-of-flight are achieved up to 0.12 a.u. and 0.02 a.u., respectively. Present photoelectron momentum distributions ionized from the ground state and a few excited states illuminate unprecedentedly rich landscapes manifesting prominent features for multi-photon absorption. The full vector momenta of electrons and recoil ion in coincidence paves the way to further studying two-electron correlation dynamics and multi-electron effects in the multiple ionization of alkaline-earth atoms in the ultraviolet region. - 11 May 2023
- Published by American Physical Society (APS) in Physical Review A
- Vol. 107 (5) , 053310
Anomalous loss behavior in a single-component Fermi gas close to a -wave Feshbach resonanceWe theoretically investigate three-body losses in a single-component Fermi gas near a -wave Feshbach resonance in the interacting, nonunitary regime. We extend the cascade model introduced by Waseem et al. [M. Waseem, J. Yoshida, T. Saito, and T. Mukaiyama, Phys. Rev. A 99, 052704 (2019)] to describe the elastic and inelastic collision processes. We find that the loss behavior exhibits a and an anomalous density dependence for a ratio of elastic-to-inelastic collision rate larger and smaller than 1, respectively. The corresponding evolutions of the energy distribution show collisional cooling or evolution toward low-energetic nonthermalized steady states, respectively. These findings are particularly relevant for understanding atom loss and the energetic evolution of ultracold gases of fermionic lithium atoms in their ground state.- 11 May 2023
- Published by American Physical Society (APS) in Physical Review A
- Vol. 107 (5) , 053310
Anomalous loss behavior in a single-component Fermi gas close to a -wave Feshbach resonanceWe theoretically investigate three-body losses in a single-component Fermi gas near a -wave Feshbach resonance in the interacting, nonunitary regime. We extend the cascade model introduced by Waseem et al. [M. Waseem, J. Yoshida, T. Saito, and T. Mukaiyama, Phys. Rev. A 99, 052704 (2019)] to describe the elastic and inelastic collision processes. We find that the loss behavior exhibits a and an anomalous density dependence for a ratio of elastic-to-inelastic collision rate larger and smaller than 1, respectively. The corresponding evolutions of the energy distribution show collisional cooling or evolution toward low-energetic nonthermalized steady states, respectively. These findings are particularly relevant for understanding atom loss and the energetic evolution of ultracold gases of fermionic lithium atoms in their ground state.- 23 March 2022
- Published by American Physical Society (APS) in Physical Review B
- Vol. 105 (10) , 104204
Dynamics of position-disordered Ising spins with a soft-core potential
We theoretically study the magnetization relaxation of Ising spins distributed randomly in a -dimension homogeneous and Gaussian profile under a soft-core two-body interaction potential , where is the interspin distance and is the soft-core radius. The dynamics starts with all spins polarized in the transverse direction. In the homogeneous case, an analytic expression is derived at the thermodynamic limit, which starts as with a constant and follows a stretched-exponential law at long time with an exponent . In between an oscillating behavior is observed with a damping amplitude. For Gaussian samples, the degree of disorder in the system can be controlled by the ratio , with the mean interspin distance and the magnetization dynamics is investigated numerically. In the limit of , a coherent many-body dynamics is recovered for the total magnetization despite the position disorder of spins. In the opposite limit of , a similar dynamics as that in the homogeneous case emerges at a later time after a initial fast decay of the magnetization. We obtain a stretched exponent of for the asymptotic evolution with , which is different from that in the homogeneous case .- 23 March 2022
- Published by American Physical Society (APS) in Physical Review B
- Vol. 105 (10) , 104204
Dynamics of position-disordered Ising spins with a soft-core potential
We theoretically study the magnetization relaxation of Ising spins distributed randomly in a -dimension homogeneous and Gaussian profile under a soft-core two-body interaction potential , where is the interspin distance and is the soft-core radius. The dynamics starts with all spins polarized in the transverse direction. In the homogeneous case, an analytic expression is derived at the thermodynamic limit, which starts as with a constant and follows a stretched-exponential law at long time with an exponent . In between an oscillating behavior is observed with a damping amplitude. For Gaussian samples, the degree of disorder in the system can be controlled by the ratio , with the mean interspin distance and the magnetization dynamics is investigated numerically. In the limit of , a coherent many-body dynamics is recovered for the total magnetization despite the position disorder of spins. In the opposite limit of , a similar dynamics as that in the homogeneous case emerges at a later time after a initial fast decay of the magnetization. We obtain a stretched exponent of for the asymptotic evolution with , which is different from that in the homogeneous case .