Showing 1–49 of 49 results for author: Hang, C

Towards a Holistic Evaluation of LLMs on Factual Knowledge Recall

Authors: Jiaqing Yuan, Lin Pan, Chung-Wei Hang, Jiang Guo, Jiarong Jiang, Bonan Min, Patrick Ng, Zhiguo Wang

Abstract: Large language models (LLMs) have shown remarkable performance on a variety of NLP tasks, and are being rapidly adopted in a wide range of use cases. It is therefore of vital importance to holistically evaluate the factuality of their generated outputs, as hallucinations remain a challenging issue. In this work, we focus on assessing LLMs' ability to recall factual knowledge learned from pretrai… ▽ More Large language models (LLMs) have shown remarkable performance on a variety of NLP tasks, and are being rapidly adopted in a wide range of use cases. It is therefore of vital importance to holistically evaluate the factuality of their generated outputs, as hallucinations remain a challenging issue. In this work, we focus on assessing LLMs' ability to recall factual knowledge learned from pretraining, and the factors that affect this ability. To that end, we construct FACT-BENCH, a representative benchmark covering 20 domains, 134 property types, 3 answer types, and different knowledge popularity levels. We benchmark 31 models from 10 model families and provide a holistic assessment of their strengths and weaknesses. We observe that instruction-tuning hurts knowledge recall, as pretraining-only models consistently outperform their instruction-tuned counterparts, and positive effects of model scaling, as larger models outperform smaller ones for all model families. However, the best performance from GPT-4 still represents a large gap with the upper-bound. We additionally study the role of in-context exemplars using counterfactual demonstrations, which lead to significant degradation of factual knowledge recall for large models. By further decoupling model known and unknown knowledge, we find the degradation is attributed to exemplars that contradict a model's known knowledge, as well as the number of such exemplars. Lastly, we fine-tune LLaMA-7B in different settings of known and unknown knowledge. In particular, fine-tuning on a model's known knowledge is beneficial, and consistently outperforms fine-tuning on unknown and mixed knowledge. We will make our benchmark publicly available. △ Less

Submitted 24 April, 2024; originally announced April 2024.

Shock wave generation and propagation in dissipative and nonlocal nonlinear Rydberg media

Authors: Lu Qin, Chao Hang, Guoxiang Huang, Weibin Li

Abstract: We investigate the generation of optical shock waves in strongly interacting Rydberg atomic gases with a spatially homogeneous dissipative potential. The Rydberg atom interaction induces an optical nonlocal nonlinarity. We focus on local nonlinear ($R_b\ll R_0$) and nonlocal nonlinear ($R_b\sim R_0$) regimes, where $R_b$ and $R_0$ are the characteristic length of the Rydberg nonlinearity and beam… ▽ More We investigate the generation of optical shock waves in strongly interacting Rydberg atomic gases with a spatially homogeneous dissipative potential. The Rydberg atom interaction induces an optical nonlocal nonlinarity. We focus on local nonlinear ($R_b\ll R_0$) and nonlocal nonlinear ($R_b\sim R_0$) regimes, where $R_b$ and $R_0$ are the characteristic length of the Rydberg nonlinearity and beam width, respectively. In the local regime, we show spatial width and contrast of the shock wave change monotonically when increasing strength of the dissipative potential and optical intensity. In the nonlocal regime, the characteristic quantity of the shock wave depend on $R_b/R_0$ and dissipative potential nontrivially and on the intensity monotonically. We find that formation of shock waves dominantly takes place when $R_b$ is smaller than $R_0$, while the propagation dynamics is largely linear when $R_b$ is comparable to or larger than $R_0$. Our results reveal nontrivial roles played by dissipation and nonlocality in the generation of shock waves, and provide a route to manipulate their profiles and stability. Our study furthermore opens new avenues to explore non-Hermitian physics, and nonlinear wave generation and propagation by controlling dissipation and nonlocality in the Rydberg media. △ Less

Submitted 9 April, 2024; originally announced April 2024.

Comments: 14 pages, 10 figures

Copilot for Xcode: Exploring AI-Assisted Programming by Prompting Cloud-based Large Language Models

Authors: Chee Wei Tan, Shangxin Guo, Man Fai Wong, Ching Nam Hang

Abstract: This paper presents an AI-assisted programming tool called Copilot for Xcode for program composition and design to support human software developers. By seamlessly integrating cloud-based Large Language Models (LLM) with Apple's local development environment, Xcode, this tool enhances productivity and unleashes creativity for software development in Apple software ecosystem (e.g., iOS apps, macOS)… ▽ More This paper presents an AI-assisted programming tool called Copilot for Xcode for program composition and design to support human software developers. By seamlessly integrating cloud-based Large Language Models (LLM) with Apple's local development environment, Xcode, this tool enhances productivity and unleashes creativity for software development in Apple software ecosystem (e.g., iOS apps, macOS). Leveraging advanced natural language processing (NLP) techniques, Copilot for Xcode effectively processes source code tokens and patterns within code repositories, enabling features such as code generation, autocompletion, documentation, and error detection. Software developers can also query and make "small" decisions for program composition, some of which can be made simultaneously, and this is facilitated through prompt engineering in a chat interface of Copilot for Xcode. Finally, we present simple case studies as evidence of the effectiveness of utilizing NLP in Xcode to prompt popular LLM services like OpenAI ChatGPT for program composition and design. △ Less

Submitted 8 July, 2023; originally announced July 2023.

Natural Language Generation and Understanding of Big Code for AI-Assisted Programming: A Review

Authors: Man Fai Wong, Shangxin Guo, Ching Nam Hang, Siu Wai Ho, Chee Wei Tan

Abstract: This paper provides a comprehensive review of the literature concerning the utilization of Natural Language Processing (NLP) techniques, with a particular focus on transformer-based large language models (LLMs) trained using Big Code, within the domain of AI-assisted programming tasks. LLMs, augmented with software naturalness, have played a crucial role in facilitating AI-assisted programming app… ▽ More This paper provides a comprehensive review of the literature concerning the utilization of Natural Language Processing (NLP) techniques, with a particular focus on transformer-based large language models (LLMs) trained using Big Code, within the domain of AI-assisted programming tasks. LLMs, augmented with software naturalness, have played a crucial role in facilitating AI-assisted programming applications, including code generation, code completion, code translation, code refinement, code summarization, defect detection, and clone detection. Notable examples of such applications include the GitHub Copilot powered by OpenAI's Codex and DeepMind AlphaCode. This paper presents an overview of the major LLMs and their applications in downstream tasks related to AI-assisted programming. Furthermore, it explores the challenges and opportunities associated with incorporating NLP techniques with software naturalness in these applications, with a discussion on extending AI-assisted programming capabilities to Apple's Xcode for mobile software development. This paper also presents the challenges of and opportunities for incorporating NLP techniques with software naturalness, empowering developers with advanced coding assistance and streamlining the software development process. △ Less

Submitted 4 July, 2023; originally announced July 2023.

Journal ref: Entropy(2023), 25(6), 888

Benchmarking Diverse-Modal Entity Linking with Generative Models

Authors: Sijia Wang, Alexander Hanbo Li, Henry Zhu, Sheng Zhang, Chung-Wei Hang, Pramuditha Perera, Jie Ma, William Wang, Zhiguo Wang, Vittorio Castelli, Bing Xiang, Patrick Ng

Abstract: Entities can be expressed in diverse formats, such as texts, images, or column names and cell values in tables. While existing entity linking (EL) models work well on per modality configuration, such as text-only EL, visual grounding, or schema linking, it is more challenging to design a unified model for diverse modality configurations. To bring various modality configurations together, we constr… ▽ More Entities can be expressed in diverse formats, such as texts, images, or column names and cell values in tables. While existing entity linking (EL) models work well on per modality configuration, such as text-only EL, visual grounding, or schema linking, it is more challenging to design a unified model for diverse modality configurations. To bring various modality configurations together, we constructed a benchmark for diverse-modal EL (DMEL) from existing EL datasets, covering all three modalities including text, image, and table. To approach the DMEL task, we proposed a generative diverse-modal model (GDMM) following a multimodal-encoder-decoder paradigm. Pre-training \Model with rich corpora builds a solid foundation for DMEL without storing the entire KB for inference. Fine-tuning GDMM builds a stronger DMEL baseline, outperforming state-of-the-art task-specific EL models by 8.51 F1 score on average. Additionally, extensive error analyses are conducted to highlight the challenges of DMEL, facilitating future research on this task. △ Less

Submitted 26 May, 2023; originally announced May 2023.

Comments: 15 pages. ACL 2023

UNITE: A Unified Benchmark for Text-to-SQL Evaluation

Authors: Wuwei Lan, Zhiguo Wang, Anuj Chauhan, Henghui Zhu, Alexander Li, Jiang Guo, Sheng Zhang, Chung-Wei Hang, Joseph Lilien, Yiqun Hu, Lin Pan, Mingwen Dong, Jun Wang, Jiarong Jiang, Stephen Ash, Vittorio Castelli, Patrick Ng, Bing Xiang

Abstract: A practical text-to-SQL system should generalize well on a wide variety of natural language questions, unseen database schemas, and novel SQL query structures. To comprehensively evaluate text-to-SQL systems, we introduce a UNIfied benchmark for Text-to-SQL Evaluation (UNITE). It is composed of publicly available text-to-SQL datasets, containing natural language questions from more than 12 domains… ▽ More A practical text-to-SQL system should generalize well on a wide variety of natural language questions, unseen database schemas, and novel SQL query structures. To comprehensively evaluate text-to-SQL systems, we introduce a UNIfied benchmark for Text-to-SQL Evaluation (UNITE). It is composed of publicly available text-to-SQL datasets, containing natural language questions from more than 12 domains, SQL queries from more than 3.9K patterns, and 29K databases. Compared to the widely used Spider benchmark, we introduce $\sim$120K additional examples and a threefold increase in SQL patterns, such as comparative and boolean questions. We conduct a systematic study of six state-of-the-art (SOTA) text-to-SQL parsers on our new benchmark and show that: 1) Codex performs surprisingly well on out-of-domain datasets; 2) specially designed decoding methods (e.g. constrained beam search) can improve performance for both in-domain and out-of-domain settings; 3) explicitly modeling the relationship between questions and schemas further improves the Seq2Seq models. More importantly, our benchmark presents key challenges towards compositional generalization and robustness issues -- which these SOTA models cannot address well. Our code and data processing script are available at https://github.com/awslabs/unified-text2sql-benchmark △ Less

Submitted 14 July, 2023; v1 submitted 25 May, 2023; originally announced May 2023.

Comments: 5 pages

Many-Body Coherence in Quantum Transport

Authors: Ching-Chi Hang, Liang-Yan Hsu

Abstract: In this study, we propose the concept of harnessing quantum coherence to control electron transport in a many-body system. Combining an open quantum system technique based on Hubbard operators, we show that many-body coherence can eliminate the well-known Coulomb staircase and cause strong negative differential resistance. To explore the mechanism, we analytically derive the current-coherence rela… ▽ More In this study, we propose the concept of harnessing quantum coherence to control electron transport in a many-body system. Combining an open quantum system technique based on Hubbard operators, we show that many-body coherence can eliminate the well-known Coulomb staircase and cause strong negative differential resistance. To explore the mechanism, we analytically derive the current-coherence relationship in the zero electron-phonon coupling limit. Furthermore, by incorporating a gate field, we demonstrate the possibility of constructing a coherence-controlled transistor. This development opens up a new direction for exploring quantum electronic devices based on many-body coherence. △ Less

Submitted 31 August, 2023; v1 submitted 21 April, 2023; originally announced April 2023.

Comments: 15 pages, 5 figures

Importance of Synthesizing High-quality Data for Text-to-SQL Parsing

Authors: Yiyun Zhao, Jiarong Jiang, Yiqun Hu, Wuwei Lan, Henry Zhu, Anuj Chauhan, Alexander Li, Lin Pan, Jun Wang, Chung-Wei Hang, Sheng Zhang, Marvin Dong, Joe Lilien, Patrick Ng, Zhiguo Wang, Vittorio Castelli, Bing Xiang

Abstract: Recently, there has been increasing interest in synthesizing data to improve downstream text-to-SQL tasks. In this paper, we first examined the existing synthesized datasets and discovered that state-of-the-art text-to-SQL algorithms did not further improve on popular benchmarks when trained with augmented synthetic data. We observed two shortcomings: illogical synthetic SQL queries from independe… ▽ More Recently, there has been increasing interest in synthesizing data to improve downstream text-to-SQL tasks. In this paper, we first examined the existing synthesized datasets and discovered that state-of-the-art text-to-SQL algorithms did not further improve on popular benchmarks when trained with augmented synthetic data. We observed two shortcomings: illogical synthetic SQL queries from independent column sampling and arbitrary table joins. To address these issues, we propose a novel synthesis framework that incorporates key relationships from schema, imposes strong ty**, and conducts schema-distance-weighted column sampling. We also adopt an intermediate representation (IR) for the SQL-to-text task to further improve the quality of the generated natural language questions. When existing powerful semantic parsers are pre-finetuned on our high-quality synthesized data, our experiments show that these models have significant accuracy boosts on popular benchmarks, including new state-of-the-art performance on Spider. △ Less

Submitted 16 December, 2022; originally announced December 2022.

Accessing and Manipulating Dispersive Shock Waves in a Nonlinear and Nonlocal Rydberg Medium

Authors: Chao Hang, Zhengyang Bai, Weibin Li, Anatoly M. Kamchatnov, Guoxiang Huang

Abstract: Dispersive shock waves (DSWs) are fascinating wave phenomena occurring in media when nonlinearity overwhelms dispersion (or diffraction). Creating DSWs with low generation power and realizing their active controls is desirable but remains a longstanding challenge. Here, we propose a scheme to generate weak-light DSWs and realize their manipulations in an atomic gas involving strongly interacting R… ▽ More Dispersive shock waves (DSWs) are fascinating wave phenomena occurring in media when nonlinearity overwhelms dispersion (or diffraction). Creating DSWs with low generation power and realizing their active controls is desirable but remains a longstanding challenge. Here, we propose a scheme to generate weak-light DSWs and realize their manipulations in an atomic gas involving strongly interacting Rydberg states under the condition of electromagnetically induced transparency (EIT). We show that for a two-dimensional (2D) Rydberg gas a weak nonlocality of optical Kerr nonlinearity can significantly change the edge speed of DSWs and induces a singular behavior of the edge speed and hence an instability of the DSWs. However, by increasing the degree of the Kerr nonlocality, the singular behavior of the edge speed and the instability of the DSWs can be suppressed. We also show that in a 3D Rydberg gas, DSWs can be created and propagate stably when the system works in the intermediate nonlocality regime. Due to the EIT effect and the giant nonlocal Kerr nonlinearity contributed by the Rydberg-Rydberg interaction, DSWs found here have extremely low generation power. In addition, an active control of DSWs can be realized; in particular, they can be stored and retrieved with high efficiency and fidelity through switching off and on a control laser field. The results reported here are useful not only for unveiling intriguing physics of DSWs but also for finding promising applications of nonlinear and nonlocal Rydberg media. △ Less

Submitted 19 October, 2022; originally announced October 2022.

Comments: 20 pages, 13 figures

Stable high-dimensional weak-light soliton molecules and their active control

Authors: Lu Qin, Chao Hang, Boris A. Malomed, Guoxiang Huang

Abstract: Bound states of solitons, alias soliton molecules (SMs), are well known in one-dimensional (1D) systems, while making stable bound states of multidimensional solitons is a challenging problem because of the underlying instabilities. Here we propose a scheme for the creation of stable (2+1)D and (3+1)D optical SMs in a gas of cold Rydberg atoms, in which electromagnetically induced transparency (EI… ▽ More Bound states of solitons, alias soliton molecules (SMs), are well known in one-dimensional (1D) systems, while making stable bound states of multidimensional solitons is a challenging problem because of the underlying instabilities. Here we propose a scheme for the creation of stable (2+1)D and (3+1)D optical SMs in a gas of cold Rydberg atoms, in which electromagnetically induced transparency (EIT) is induced by a control laser field. We show that, through the interplay of the EIT and the strong long-range interaction between the Rydberg atoms, the system gives rise to giant nonlocal Kerr nonlinearity, which in turn supports stable (2+1)D spatial optical SMs, as well as ring-shaped soliton necklaces, including rotating ones. They feature a large size, low generation power, and can be efficiently manipulated by tuning the nonlocality degree of the Kerr nonlinearity. Stable (3+1)D spatiotemporal optical SMs, composed of fundamental or vortex solitons, with low power and ultraslow propagation velocity, can also be generated in the system. These SMs can be stored and retrieved through the switching off and on of the control laser field. The findings reported here suggest applications to data processing and transmission in optical systems. △ Less

Submitted 14 July, 2022; v1 submitted 5 June, 2022; originally announced June 2022.

Comments: The manuscript has been accepted by laser & Photonics Reviews for publication

Influence of perspective taking through robotic virtual agents on prosocial behavior

Authors: Chenlin Hang, Tetsuo Ono, Seiji Yamada

Abstract: Perspective taking, which allows people to imagine another's thinking and goals, is known to be an effective method for promoting prosocial behaviors in human-computer interactions. However, most of the previous studies have focused on simulating human-human interactions in the real world by offering participants experiences related to various moral tasks through the use of human-like virtual agen… ▽ More Perspective taking, which allows people to imagine another's thinking and goals, is known to be an effective method for promoting prosocial behaviors in human-computer interactions. However, most of the previous studies have focused on simulating human-human interactions in the real world by offering participants experiences related to various moral tasks through the use of human-like virtual agents. In this study, we investigated whether taking the perspective of a different robot in a robot-altruistic task would influence the social behaviors of participants in a dictator game. Our findings showed that participants who watched the help-receiver view exhibited more altruistic behaviors toward a robot than those who watched the help-provider view. We also found that, after watching robots from two different viewpoints in the task, participants did not change their behavior toward another participant. △ Less

Submitted 24 May, 2022; originally announced May 2022.

Designing nudge agents that promote human altruism

Authors: Chenlin Hang, Tetsuo Ono, Seiji Yamada

Abstract: Previous studies have found that nudging is key to promoting altruism in human-human interaction. However, in social robotics, there is still a lack of study on confirming the effect of nudging on altruism. In this paper, we apply two nudge mechanisms, peak-end and multiple viewpoints, to a video stimulus performed by social robots (virtual agents) to see whether a subtle change in the stimulus ca… ▽ More Previous studies have found that nudging is key to promoting altruism in human-human interaction. However, in social robotics, there is still a lack of study on confirming the effect of nudging on altruism. In this paper, we apply two nudge mechanisms, peak-end and multiple viewpoints, to a video stimulus performed by social robots (virtual agents) to see whether a subtle change in the stimulus can promote human altruism. An experiment was conducted online through crowdsourcing with 136 participants. The result shows that the participants who watched the peak part set at the end of the video performed better at the Dictator game, which means that the nudge mechanism of the peak-end effect actually promoted human altruism. △ Less

Submitted 1 October, 2021; originally announced October 2021.

Comments: 11 pages, 5 figures, accepted by The Thirteenth International Conference on Social Robotics

Experimental Observation of Partial Parity-Time Symmetry and Its Phase Transition with a Laser-Driven Cesium Atomic Gas

Authors: Yongmei Xue, Chao Hang, Yunhui He, Zhengyang Bai, Yuechun Jiao, Guoxiang Huang, Jianming Zhao, Suotang Jia

Abstract: Realization and manipulation of parity-time (PT) symmetry in multidimensional systems are highly desirable for exploring nontrivial physics and uncovering exotic phenomena in non-Hermitian systems. Here, we report the first experimental observation of partial PT (pPT) symmetry in a cesium atomic gas coupled with laser fields, where a two-dimensional pPT-symmetric optical potential for probe laser… ▽ More Realization and manipulation of parity-time (PT) symmetry in multidimensional systems are highly desirable for exploring nontrivial physics and uncovering exotic phenomena in non-Hermitian systems. Here, we report the first experimental observation of partial PT (pPT) symmetry in a cesium atomic gas coupled with laser fields, where a two-dimensional pPT-symmetric optical potential for probe laser beam is created. A transition of the pPT symmetry from an unbroken phase to a broken one is observed through changing the beam-waist ratio of the control and probe laser beams, and the domains of unbroken, broken, and non-pPT phases are also discriminated unambiguously. Moreover, we develop a technique to precisely determine the location of the exceptional point of the pPT symmetry breaking by measuring the asymmetry degree of the probe-beam intensity distribution. The findings reported here pave the way for controlling multidimensional laser beams in non-Hermitian systems via laser-induced atomic coherence, and have potential applications for designing new types of light amplifiers and attenuators △ Less

Submitted 19 September, 2021; originally announced September 2021.

Comments: 6 pages (including references), 4 figures

Improved Text Classification via Contrastive Adversarial Training

Authors: Lin Pan, Chung-Wei Hang, Avirup Sil, Saloni Potdar

Abstract: We propose a simple and general method to regularize the fine-tuning of Transformer-based encoders for text classification tasks. Specifically, during fine-tuning we generate adversarial examples by perturbing the word embeddings of the model and perform contrastive learning on clean and adversarial examples in order to teach the model to learn noise-invariant representations. By training on both… ▽ More We propose a simple and general method to regularize the fine-tuning of Transformer-based encoders for text classification tasks. Specifically, during fine-tuning we generate adversarial examples by perturbing the word embeddings of the model and perform contrastive learning on clean and adversarial examples in order to teach the model to learn noise-invariant representations. By training on both clean and adversarial examples along with the additional contrastive objective, we observe consistent improvement over standard fine-tuning on clean examples. On several GLUE benchmark tasks, our fine-tuned BERT Large model outperforms BERT Large baseline by 1.7% on average, and our fine-tuned RoBERTa Large improves over RoBERTa Large baseline by 1.3%. We additionally validate our method in different domains using three intent classification datasets, where our fine-tuned RoBERTa Large outperforms RoBERTa Large baseline by 1-2% on average. △ Less

Submitted 17 February, 2022; v1 submitted 21 July, 2021; originally announced July 2021.

Nonlinear Topological Edge States in a non-Hermitian Array of Optical Waveguides Embedded in an Atomic Gas

Authors: Chao Hang, Dmitry A. Zezyulin, Guoxiang Huang, Vladimir V. Konotop

Abstract: We propose a scheme comprising an array of anisotropic optical waveguides, embedded in a gas of cold atoms, which can be tuned from a Hermitian to an odd-PT -- symmetric configuration through the manipulation of control and assistant laser fields. We show that the system can be controlled by tuning intra -- and inter-cell coupling coefficients, enabling the creation of topologically distinct phase… ▽ More We propose a scheme comprising an array of anisotropic optical waveguides, embedded in a gas of cold atoms, which can be tuned from a Hermitian to an odd-PT -- symmetric configuration through the manipulation of control and assistant laser fields. We show that the system can be controlled by tuning intra -- and inter-cell coupling coefficients, enabling the creation of topologically distinct phases and linear topological edge states. The waveguide array, characterized by a quadrimer primitive cell, allows for implementing transitions between Hermitian and odd-PT -symmetric configurations, broken and unbroken PT -symmetric phases, topologically trivial and nontrivial phases, as well as transitions between linear and nonlinear regimes. The introduced scheme generalizes the Rice-Mele Hamiltonian for a nonlinear non-Hermitian quadrimer array featuring odd-PT symmetry and makes accessible unique phenomena and functionalities that emerge from the interplay of non-Hermiticity, topology, and nonlinearity. We also show that in the presence of nonlinearity the system sustains nonlinear topological edge states bifurcating from the linear topological edge states and the modes without linear limit. Each nonlinear mode represents a doublet of odd-PT -conjugate states. In the broken PT phase, the nonlinear edge states may be effectively stabilized when an additional absorption is introduced into the system. △ Less

Submitted 3 April, 2021; originally announced April 2021.

Comments: 7 pages, 3 figures

Journal ref: Physical Review A 103, L040202 (2021)

Multilingual BERT Post-Pretraining Alignment

Authors: Lin Pan, Chung-Wei Hang, Haode Qi, Abhishek Shah, Saloni Potdar, Mo Yu

Abstract: We propose a simple method to align multilingual contextual embeddings as a post-pretraining step for improved zero-shot cross-lingual transferability of the pretrained models. Using parallel data, our method aligns embeddings on the word level through the recently proposed Translation Language Modeling objective as well as on the sentence level via contrastive learning and random input shuffling.… ▽ More We propose a simple method to align multilingual contextual embeddings as a post-pretraining step for improved zero-shot cross-lingual transferability of the pretrained models. Using parallel data, our method aligns embeddings on the word level through the recently proposed Translation Language Modeling objective as well as on the sentence level via contrastive learning and random input shuffling. We also perform sentence-level code-switching with English when finetuning on downstream tasks. On XNLI, our best model (initialized from mBERT) improves over mBERT by 4.7% in the zero-shot setting and achieves comparable result to XLM for translate-train while using less than 18% of the same parallel data and 31% less model parameters. On MLQA, our model outperforms XLM-R_Base that has 57% more parameters than ours. △ Less

Submitted 10 April, 2021; v1 submitted 23 October, 2020; originally announced October 2020.

Comments: Accepted at NAACL2021

Numerical simulations and Arctic observations of surface wind effects on Multi-Angle Snowflake Camera measurements

Authors: Kyle E. Fitch, Chaoxun Hang, Ahmad Talaei, Timothy J. Garrett

Abstract: Ground-based measurements of frozen precipitation are heavily influenced by interactions of surface winds with gauge-shield geometry. The Multi-Angle Snowflake Camera (MASC), which photographs hydrometeors in free-fall from three different angles while simultaneously measuring their fall speed, has been used in the field at multiple mid-latitude and polar locations both with and without wind shiel… ▽ More Ground-based measurements of frozen precipitation are heavily influenced by interactions of surface winds with gauge-shield geometry. The Multi-Angle Snowflake Camera (MASC), which photographs hydrometeors in free-fall from three different angles while simultaneously measuring their fall speed, has been used in the field at multiple mid-latitude and polar locations both with and without wind shielding. Here we show results of computational fluid dynamics (CFD) simulations of the airflow and corresponding particle trajectories around the unshielded MASC and compare these results to Arctic field observations with and without a Belfort double Alter shield. Simulations in the absence of a wind shield show a separation of flow at the upstream side of the instrument, with an upward velocity component just above the aperture, which decreases the mean particle fall speed by 55%(74%) for a wind speed of 5 m/s(10 m/s). MASC-measured fall speeds compare well with Ka-band Atmospheric Radiation Measurement (ARM) Zenith Radar (KAZR) mean Doppler velocities only when winds are light (<5 m/s) and the MASC is shielded. MASC-measured fall speeds that do not match KAZR measured velocities tend to fall below a threshold value that increases approximately linearly with wind speed but is generally <0.5 m/s. For those events with wind speeds <1.5 m/s, hydrometeors fall with an orientation angle mode of 12 degrees from the horizontal plane, and large, low-density aggregates are as much as five times more likely to be observed. We conclude that accurate MASC observations of the microphysical, orientation, and fall speed characteristics of snow particles require shielding by a double wind fence and restriction of analysis to events where winds are light (<5 m/s). Hydrometeors do not generally fall in still air, so adjustments to these properties' distributions within natural turbulence remain to be determined. △ Less

Submitted 21 July, 2020; originally announced July 2020.

Comments: 26 pages including 15 figures and 3 tables

Octa: Omissions and Conflicts in Target-Aspect Sentiment Analysis

Authors: Zhe Zhang, Chung-Wei Hang, Munindar P. Singh

Abstract: Sentiments in opinionated text are often determined by both aspects and target words (or targets). We observe that targets and aspects interrelate in subtle ways, often yielding conflicting sentiments. Thus, a naive aggregation of sentiments from aspects and targets treated separately, as in existing sentiment analysis models, impairs performance. We propose Octa, an approach that jointly consid… ▽ More Sentiments in opinionated text are often determined by both aspects and target words (or targets). We observe that targets and aspects interrelate in subtle ways, often yielding conflicting sentiments. Thus, a naive aggregation of sentiments from aspects and targets treated separately, as in existing sentiment analysis models, impairs performance. We propose Octa, an approach that jointly considers aspects and targets when inferring sentiments. To capture and quantify relationships between targets and context words, Octa uses a selective self-attention mechanism that handles implicit or missing targets. Specifically, Octa involves two layers of attention mechanisms for, respectively, selective attention between targets and context words and attention over words based on aspects. On benchmark datasets, Octa outperforms leading models by a large margin, yielding (absolute) gains in accuracy of 1.6% to 4.3%. △ Less

Submitted 6 October, 2020; v1 submitted 27 April, 2020; originally announced April 2020.

Comments: Accepted by Findings of EMNLP 2020

Adaptive Control for Marine Vessels Against Harsh Environmental Variation

Authors: Fangwen Tu, Shuzhi Sam Ge, Yoo Sang Choo, Chang Chieh Hang

Abstract: In this paper, robust control with sea state observer and dynamic thrust allocation is proposed for the Dynamic Positioning (DP) of an accommodation vessel in the presence of unknown hydrodynamic force variation and the input time delay. In order to overcome the huge force variation due to the adjoining Floating Production Storage and Offloading (FPSO) and accommodation vessel, a novel sea state o… ▽ More In this paper, robust control with sea state observer and dynamic thrust allocation is proposed for the Dynamic Positioning (DP) of an accommodation vessel in the presence of unknown hydrodynamic force variation and the input time delay. In order to overcome the huge force variation due to the adjoining Floating Production Storage and Offloading (FPSO) and accommodation vessel, a novel sea state observer is designed. The sea observer can effectively monitor the variation of the drift wave-induced force on the vessel and activate Neural Network (NN) compensator in the controller when large wave force is identified. Moreover, the wind drag coefficients can be adaptively approximated in the sea observer so that a feedforward control can be achieved. Based on this, a robust constrained control is developed to guarantee a safe operation. The time delay inside the control input is also considered. Dynamic thrust allocation module is presented to distribute the generalized control input among azimuth thrusters. Under the proposed sea observer and control, the boundedness of all the closed-loop signals are demonstrated via rigorous Lyapunov analysis. A set of simulation studies are conducted to verify the effectiveness of the proposed control scheme. △ Less

Submitted 29 September, 2019; originally announced September 2019.

Saliency Guided Hierarchical Robust Visual Tracking

Authors: Fangwen Tu, Shuzhi Sam Ge, Yazhe Tang, Chang Chieh Hang

Abstract: A saliency guided hierarchical visual tracking (SHT) algorithm containing global and local search phases is proposed in this paper. In global search, a top-down saliency model is novelly developed to handle abrupt motion and appearance variation problems. Nineteen feature maps are extracted first and combined with online learnt weights to produce the final saliency map and estimated target locatio… ▽ More A saliency guided hierarchical visual tracking (SHT) algorithm containing global and local search phases is proposed in this paper. In global search, a top-down saliency model is novelly developed to handle abrupt motion and appearance variation problems. Nineteen feature maps are extracted first and combined with online learnt weights to produce the final saliency map and estimated target locations. After the evaluation of integration mechanism, the optimum candidate patch is passed to the local search. In local search, a superpixel based HSV histogram matching is performed jointly with an L2-RLS tracker to take both color distribution and holistic appearance feature of the object into consideration. Furthermore, a linear refinement search process with fast iterative solver is implemented to attenuate the possible negative influence of dominant particles. Both qualitative and quantitative experiments are conducted on a series of challenging image sequences. The superior performance of the proposed method over other state-of-the-art algorithms is demonstrated by comparative study. △ Less

Submitted 21 December, 2018; originally announced December 2018.

Shallow Cue Guided Deep Visual Tracking via Mixed Models

Authors: Fangwen Tu, Shuzhi Sam Ge, Chang Chieh Hang

Abstract: In this paper, a robust visual tracking approach via mixed model based convolutional neural networks (SDT) is developed. In order to handle abrupt or fast motion, a prior map is generated to facilitate the localization of region of interest (ROI) before the deep tracker is performed. A top-down saliency model with nineteen shallow cues are employed to construct the prior map with online learnt com… ▽ More In this paper, a robust visual tracking approach via mixed model based convolutional neural networks (SDT) is developed. In order to handle abrupt or fast motion, a prior map is generated to facilitate the localization of region of interest (ROI) before the deep tracker is performed. A top-down saliency model with nineteen shallow cues are employed to construct the prior map with online learnt combination weights. Moreover, apart from a holistic deep learner, four local networks are also trained to learn different components of the target. The generated four local heat maps will facilitate to rectify the holistic map by eliminating the distracters to avoid drifting. Furthermore, to guarantee the instance for online update of high quality, a prioritised update strategy is implemented by casting the problem into a label noise problem. The selection probability is designed by considering both confidence values and bio-inspired memory for temporal information integration. Experiments are conducted qualitatively and quantitatively on a set of challenging image sequences. Comparative study demonstrates that the proposed algorithm outperforms other state-of-the-art methods. △ Less

Submitted 19 December, 2018; originally announced December 2018.

Low-loss Airy Surface Plasmon Polaritons

Authors: Qi Zhang, Chaohua Tan, Chao Hang, Guoxiang Huang

Abstract: We propose a scheme to obtain a low-loss propagation of Airy surface plasmon polaritons (SPPs) along the interface between a dielectric and a negative-index metamaterial (NIMM). We show that, by using the transverse-magnetic mode and the related destructive interference effect between electric and magnetic absorption responses, the propagation loss of the Airy SPPs can be largely suppressed when t… ▽ More We propose a scheme to obtain a low-loss propagation of Airy surface plasmon polaritons (SPPs) along the interface between a dielectric and a negative-index metamaterial (NIMM). We show that, by using the transverse-magnetic mode and the related destructive interference effect between electric and magnetic absorption responses, the propagation loss of the Airy SPPs can be largely suppressed when the optical frequency is close to the lossless point of the NIMM. As a result, the Airy SPPs obtained in our scheme can propagate more than 6-time long distance than that in conventional dielectric-metal interfaces. △ Less

Submitted 31 March, 2018; originally announced April 2018.

Journal ref: Chin. Opt. Lett. 13, 082401 (2015)

The Causal Link between News Framing and Legislation

Authors: Karthik Sheshadri, Chung-Wei Hang, Munindar Singh

Abstract: We demonstrate that framing, a subjective aspect of news, is a causal precursor to both significant public perception changes, and to federal legislation. We posit, counter-intuitively, that topic news volume and mean article similarity increase and decrease together. We show that specific features of news, such as publishing volume , are predictive of both sustained public attention, measured by… ▽ More We demonstrate that framing, a subjective aspect of news, is a causal precursor to both significant public perception changes, and to federal legislation. We posit, counter-intuitively, that topic news volume and mean article similarity increase and decrease together. We show that specific features of news, such as publishing volume , are predictive of both sustained public attention, measured by annual Google trend data, and federal legislation. We observe that public attention changes are driven primarily by periods of high news volume and mean similarity, which we call \emph{prenatal periods}. Finally, we demonstrate that framing during prenatal periods may be characterized by high-utility news \emph{keywords}. △ Less

Submitted 15 February, 2018; originally announced February 2018.

Framing Matters: Predicting Framing Changes and Legislation from Topic News Patterns

Authors: Karthik Sheshadri, Chung-Wei Hang, Munindar Singh

Abstract: News has traditionally been well researched, with studies ranging from sentiment analysis to event detection and topic tracking. We extend the focus to two surprisingly under-researched aspects of news: \emph{framing} and \emph{predictive utility}. We demonstrate that framing influences public opinion and behavior, and present a simple entropic algorithm to characterize and detect framing changes.… ▽ More News has traditionally been well researched, with studies ranging from sentiment analysis to event detection and topic tracking. We extend the focus to two surprisingly under-researched aspects of news: \emph{framing} and \emph{predictive utility}. We demonstrate that framing influences public opinion and behavior, and present a simple entropic algorithm to characterize and detect framing changes. We introduce a dataset of news topics with framing changes, harvested from manual surveys in previous research. Our approach achieves an F-measure of $F_1=0.96$ on our data, whereas dynamic topic modeling returns $F_1=0.1$. We also establish that news has \emph{predictive utility}, by showing that legislation in topics of current interest can be foreshadowed and predicted from news patterns. △ Less

Submitted 15 February, 2018; originally announced February 2018.

Soliton Diffusion as a Signature of Hawking Radiation in Bose-Einstein Condensates

Authors: Chao Hang, Gregory Gabadadze, Guoxiang Huang

Abstract: We propose a scheme to detect analog Hawking radiation (HR) in an atomic Bose-Einstein condensate (BEC) through measuring the diffusion of a dark soliton. The HR is generated by changing the transverse trap** potential of the BEC to obtain a background flow, which is subsonic in downstream and supersonic in upstream, satisfying the condition of black hole horizon. When the system is in thermal e… ▽ More We propose a scheme to detect analog Hawking radiation (HR) in an atomic Bose-Einstein condensate (BEC) through measuring the diffusion of a dark soliton. The HR is generated by changing the transverse trap** potential of the BEC to obtain a background flow, which is subsonic in downstream and supersonic in upstream, satisfying the condition of black hole horizon. When the system is in thermal equilibrium at Hawking temperature, a dark soliton is created in the upstream. Due to the influence of the HR, the motion of the dark soliton is similar to a Brownian particle and hence exhibits an apparent diffusion, which can be measured and be taken as a signal of the HR. Since the dark soliton is much "heavier" than Hawking quanta, its diffusion is much easier detectable than the Hawking quanta themselves. △ Less

Submitted 28 July, 2019; v1 submitted 27 December, 2017; originally announced December 2017.

Comments: 8 pages, 3 figures

Journal ref: Physics Letters B, Volume 793, Pages 390-395 (2019)

Hamilton path decompositions of complete multipartite graphs

Authors: Darryn Bryant, Hao Chuien Hang, Sarada Herke

Abstract: We prove that a complete multipartite graph $K$ with $n>1$ vertices and $m$ edges can be decomposed into edge-disjoint Hamilton paths if and only if $\frac m{n-1}$ is an integer and the maximum degree of $K$ is at most $\frac {2m}{n-1}$. We prove that a complete multipartite graph $K$ with $n>1$ vertices and $m$ edges can be decomposed into edge-disjoint Hamilton paths if and only if $\frac m{n-1}$ is an integer and the maximum degree of $K$ is at most $\frac {2m}{n-1}$. △ Less

Submitted 23 July, 2018; v1 submitted 12 October, 2017; originally announced October 2017.

MSC Class: 05C51; 05C45

Multi-path photon-phonon converter in optomechanical system at single-quantum level

Authors: Tian-Yi Chen, Wen-Zhao Zhang, Ren-Zhou Fang, Cheng-Zhou Hang, Ling Zhou

Abstract: Based on photon-phonon nonlinear interaction, a scheme is proposed to realize a controllable multi-path photon-phonon converter at single-quantum level in a composed quadratically coupled optomechanical system. Considering the realization of the scheme, an associated mechanical oscillator is introduced to enhance the effective nonlinear effect. Thus, the single-photon state can be converted to the… ▽ More Based on photon-phonon nonlinear interaction, a scheme is proposed to realize a controllable multi-path photon-phonon converter at single-quantum level in a composed quadratically coupled optomechanical system. Considering the realization of the scheme, an associated mechanical oscillator is introduced to enhance the effective nonlinear effect. Thus, the single-photon state can be converted to the phonon state with high fidelity even under the current experimental condition that the single-photon coupling rate is much smaller than mechanical frequency ($g\llω_m$). The state transfer protocols and their transfer fidelity are discussed both analytically and numerically. A multi-path photon-phonon converter is designed, by combining the optomechanical system with low frequency resonators, which can be controlled by experimentally adjustable parameters. This work provides us a potential platform for quantum state transfer and quantum information processing. △ Less

Submitted 19 January, 2017; originally announced January 2017.

Three-dimensional topological solitons in PT-symmetric optical lattices

Authors: Yaroslav V. Kartashov, Chao Hang, Guoxiang Huang, Lluis Torner

Abstract: We address the properties of fully three-dimensional solitons in complex parity-time (PT)-symmetric periodic lattices with focusing Kerr nonlinearity, and uncover that such lattices can stabilize both, fundamental and vortex-carrying soliton states. The imaginary part of the lattice induces internal currents in the solitons that strongly affect their domains of existence and stability. The domain… ▽ More We address the properties of fully three-dimensional solitons in complex parity-time (PT)-symmetric periodic lattices with focusing Kerr nonlinearity, and uncover that such lattices can stabilize both, fundamental and vortex-carrying soliton states. The imaginary part of the lattice induces internal currents in the solitons that strongly affect their domains of existence and stability. The domain of stability for fundamental solitons can extend nearly up to the PT-symmetry breaking point, where the linear lattice spectrum becomes complex. Vortex solitons feature spatially asymmetric profiles in the PT-symmetric lattices, but they are found to still exist as stable states within narrow regions. Our results provide the first example of continuous families of stable three-dimensional propagating solitons supported by complex potentials. △ Less

Submitted 27 August, 2016; v1 submitted 23 August, 2016; originally announced August 2016.

Comments: 7 pages, 5 figures, to appear in Optica

Journal ref: Optica 3, 1048 (2016)

Storage and retrieval of (3+1)-dimensional weak-light bullets and vortices in a coherent atomic gas

Authors: Zhiming Chen, Zhengyang Bai, Hui-jun Li, Chao Hang, Guoxiang Huang

Abstract: A robust light storage and retrieval (LSR) in high dimensions is highly desirable for light and quantum information processing. However, most schemes on LSR realized up to now encounter problems due to not only dissipation, but also dispersion and diffraction, which make LSR with a very low fidelity. Here we propose a scheme to achieve a robust storage and retrieval of weak nonlinear high-dimensio… ▽ More A robust light storage and retrieval (LSR) in high dimensions is highly desirable for light and quantum information processing. However, most schemes on LSR realized up to now encounter problems due to not only dissipation, but also dispersion and diffraction, which make LSR with a very low fidelity. Here we propose a scheme to achieve a robust storage and retrieval of weak nonlinear high-dimensional light pulses in a coherent atomic gas via electromagnetically induced transparency. We show that it is available to produce stable (3+1)-dimensional light bullets and vortices, which have very attractive physical property and are suitable to obtain a robust LSR in high dimensions. △ Less

Submitted 18 April, 2016; originally announced April 2016.

Suppression and restoration of disorder-induced light localization mediated by PT-symmetry breaking

Authors: Yaroslav V. Kartashov, Chao Hang, Vladimir V. Konotop, Victor A. Vysloukh, Guoxiang Huang, Lluis Torner

Abstract: We uncover that the breaking point of the PT-symmetry in optical waveguide arrays has a dramatic impact on light localization induced by the off-diagonal disorder. Specifically, when the gain/loss control parameter approaches a critical value at which PT-symmetry breaking occurs, a fast growth of the coupling between neighboring waveguides causes diffraction to dominate to an extent that light loc… ▽ More We uncover that the breaking point of the PT-symmetry in optical waveguide arrays has a dramatic impact on light localization induced by the off-diagonal disorder. Specifically, when the gain/loss control parameter approaches a critical value at which PT-symmetry breaking occurs, a fast growth of the coupling between neighboring waveguides causes diffraction to dominate to an extent that light localization is strongly suppressed and statistically averaged width of the output pattern substantially increases. Beyond the symmetry-breaking point localization is gradually restored, although in this regime the power of localized modes grows upon propagation. The strength of localization monotonically increases with disorder at both, broken and unbroken PT-symmetry. △ Less

Submitted 8 December, 2015; originally announced December 2015.

Comments: 14 pages, 6 figures, to appear in Laser and Photonics Reviews

Journal ref: Laser and Photonics Reviews 10, 100-107 (2016)

Empirical Study on Deep Learning Models for Question Answering

Authors: Yang Yu, Wei Zhang, Chung-Wei Hang, Bing Xiang, Bowen Zhou

Abstract: In this paper we explore deep learning models with memory component or attention mechanism for question answering task. We combine and compare three models, Neural Machine Translation, Neural Turing Machine, and Memory Networks for a simulated QA data set. This paper is the first one that uses Neural Machine Translation and Neural Turing Machines for solving QA tasks. Our results suggest that the… ▽ More In this paper we explore deep learning models with memory component or attention mechanism for question answering task. We combine and compare three models, Neural Machine Translation, Neural Turing Machine, and Memory Networks for a simulated QA data set. This paper is the first one that uses Neural Machine Translation and Neural Turing Machines for solving QA tasks. Our results suggest that the combination of attention and memory have potential to solve certain QA problem. △ Less

Submitted 20 November, 2015; v1 submitted 26 October, 2015; originally announced October 2015.

Spatial solitons and stability in self-focusing and defocusing Kerr nonlinear media with generalized PT-symmetric Scarff-II potentials

Authors: Zhenya Yan, Zichao Wen, Chao Hang

Abstract: We present a unified theoretical study of the bright solitons governed by self-focusing and defocusing nonlinear Schrodinger (NLS) equations with generalized parity-time (PT)-symmetric Scarff II potentials. Particularly, a PT-symmetric k-wavenumber Scarff II potential and a multi-well Scarff II potential are considered, respectively. For the k-wavenumber Scarff II potential, the parameter space ca… ▽ More We present a unified theoretical study of the bright solitons governed by self-focusing and defocusing nonlinear Schrodinger (NLS) equations with generalized parity-time (PT)-symmetric Scarff II potentials. Particularly, a PT-symmetric k-wavenumber Scarff II potential and a multi-well Scarff II potential are considered, respectively. For the k-wavenumber Scarff II potential, the parameter space can be divided into different regions, corresponding to unbroken and broken PT-symmetry and the bright solitons for self-focusing and defocusing Kerr nonlinearities. For the multi-well Scarff II potential the bright solitons can be obtained by using a periodically space-modulated Kerr nonlinearity. The linear stability of bright solitons with PT-symmetric k-wavenumber and multi-well Scarff II potentials is analyzed in details using numerical simulations. Stable and unstable bright solitons are found in both regions of unbroken and broken PT-symmetry due to the existence of the nonlinearity. Furthermore, the bright solitons in three-dimensional self-focusing and defocusing NLS equations with a generalized PT-symmetric Scarff II potential are explored. This may have potential applications in the field of optical information transmission and processing based on optical solitons in nonlinear dissipative but PT-symmetric systems. △ Less

Submitted 19 September, 2015; originally announced September 2015.

Comments: 11 pages, 13 figures

Journal ref: Phys. Rev. E 92, 022913 (2015)

PT-symmetric coupler with a coupling defect: soliton interaction with exceptional point

Authors: Yuli V. Bludov, Chao Hang, Guoxiang Huang, Vladimir V. Konotop

Abstract: We study interaction of a soliton in a parity-time (PT) symmetric coupler which has local perturbation of the coupling constant. Such a defect does not change the PT-symmetry of the system, but locally can achieve the exceptional point. We found that the symmetric solitons after interaction with the defect either transform into breathers or blow up. The dynamics of anti-symmetric solitons is more… ▽ More We study interaction of a soliton in a parity-time (PT) symmetric coupler which has local perturbation of the coupling constant. Such a defect does not change the PT-symmetry of the system, but locally can achieve the exceptional point. We found that the symmetric solitons after interaction with the defect either transform into breathers or blow up. The dynamics of anti-symmetric solitons is more complex, showing domains of successive broadening of the beam and of the beam splitting in two outwards propagating solitons, in addition to the single breather generation and blow up. All the effects are preserved when the coupling strength in the center of the defect deviates from the exceptional point. If the coupling is strong enough the only observable outcome of the soliton-defect interaction is the generation of the breather. △ Less

Submitted 8 May, 2014; originally announced May 2014.

Comments: 4 pages, 6 figures; accepted by Opt. Lett

Guiding ultraslow weak-light bullets with Airy beams in a coherent atomic system

Authors: Chao Hang, Guoxiang Huang

Abstract: We investigate the possibility of guiding stable ultraslow weak-light bullets by using Airy beams in a cold, lifetime-broadened four-level atomic system via electromagnetically induced transparency (EIT). We show that under EIT condition the light bullet with ultraslow propagating velocity and extremely low generation power formed by the balance between diffraction and nonlinearity in the probe fi… ▽ More We investigate the possibility of guiding stable ultraslow weak-light bullets by using Airy beams in a cold, lifetime-broadened four-level atomic system via electromagnetically induced transparency (EIT). We show that under EIT condition the light bullet with ultraslow propagating velocity and extremely low generation power formed by the balance between diffraction and nonlinearity in the probe field can be not only stabilized but also steered by the assisted field. In particular, when the assisted field is taken to be an Airy beam, the light bullet can be trapped into the main lobe of the Airy beam, propagate ultraslowly in longitudinal direction, accelerate in transverse directions, and move along a parabolic trajectory. We further show that the light bullet can bypass an obstacle when guided by two sequential Airy beams. A technique for generating ultraslow helical weak-light bullets is also proposed. △ Less

Submitted 8 May, 2014; originally announced May 2014.

Comments: 23 pages, 6 figures

Journal ref: Phys. Rev. A 89, 013821 (2014)

Ultraslow Helical Optical Bullets and Their Acceleration in Magneto-Optically Controlled Coherent Atomic Media

Authors: Chao Hang, Guoxiang Huang

Abstract: We propose a scheme to produce ultraslow (3+1)-dimensional helical optical solitons, alias helical optical bullets, in a resonant three-level $Λ$-type atomic system via quantum coherence. We show that, due to the effect of electromagnetically induced transparency, the helical optical bullets can propagate with an ultraslow velocity up to $10^{-5}$ $c$ ($c$ is the light speed in vacuum) in longitud… ▽ More We propose a scheme to produce ultraslow (3+1)-dimensional helical optical solitons, alias helical optical bullets, in a resonant three-level $Λ$-type atomic system via quantum coherence. We show that, due to the effect of electromagnetically induced transparency, the helical optical bullets can propagate with an ultraslow velocity up to $10^{-5}$ $c$ ($c$ is the light speed in vacuum) in longitudinal direction and a slow rotational motion (with velocity $10^{-7}$ $c$) in transverse directions. The generation power of such optical bullets can be lowered to microwatt, and their stability can be achieved by using a Bessel optical lattice potential formed by a far-detuned laser field. We also show that the transverse rotational motion of the optical bullets can be accelerated by applying a time-dependent Stern-Gerlach magnetic field. Because of the untraslow velocity in the longitudinal direction, a significant acceleration of the rotational motion of optical bullets may be observed for a very short medium length. △ Less

Submitted 8 April, 2014; originally announced April 2014.

Comments: 21 pages, 5 figures

Journal ref: Phys. Rev. A 87, 053809 (2013)

Slow-light Airy wave packets and their active control via electromagnetically induced transparency

Authors: Chao Hang, Guoxiang Huang

Abstract: We propose a scheme to generate (3+1)-dimensional slow-light Airy wave packets in a resonant $Λ$-type three-level atomic gas via electromagnetically induced transparency. We show that in the absence of dispersion the Airy wave packets formed by a probe field consist of two Airy wave packets accelerated in transverse directions and a longitudinal Gaussian pulse with a constant propagating velocity… ▽ More We propose a scheme to generate (3+1)-dimensional slow-light Airy wave packets in a resonant $Λ$-type three-level atomic gas via electromagnetically induced transparency. We show that in the absence of dispersion the Airy wave packets formed by a probe field consist of two Airy wave packets accelerated in transverse directions and a longitudinal Gaussian pulse with a constant propagating velocity lowered to $10^{-5}\,c$ ($c$ is the light speed in vacuum). We also show that in the presence of dispersion it is possible to generate another type of slow-light Airy wave packets consisting of two Airy beams in transverse directions and an Airy wave packet in the longitudinal direction. In this case, the longitudinal velocity of the Airy wave packet can be further reduced during propagation. Additionally, we further show that the transverse accelerations (or bending) of the both types of slow-light Airy wave packets can be completely eliminated and the motional trajectories of them can be actively manipulated and controlled by using a Stern-Gerlach gradient magnetic field. △ Less

Submitted 8 April, 2014; originally announced April 2014.

Comments: 19 pages, 5 figures

Journal ref: Phys. Rev. A 88, 013825 (2013)

A Probabilistic Approach for Maintaining Trust Based on Evidence

Authors: Yonghong Wang, Chung-Wei Hang, Munindar P. Singh

Abstract: Leading agent-based trust models address two important needs. First, they show how an agent may estimate the trustworthiness of another agent based on prior interactions. Second, they show how agents may share their knowledge in order to cooperatively assess the trustworthiness of others. However, in real-life settings, information relevant to trust is usually obtained piecemeal, not all at onc… ▽ More Leading agent-based trust models address two important needs. First, they show how an agent may estimate the trustworthiness of another agent based on prior interactions. Second, they show how agents may share their knowledge in order to cooperatively assess the trustworthiness of others. However, in real-life settings, information relevant to trust is usually obtained piecemeal, not all at once. Unfortunately, the problem of maintaining trust has drawn little attention. Existing approaches handle trust updates in a heuristic, not a principled, manner. This paper builds on a formal model that considers probability and certainty as two dimensions of trust. It proposes a mechanism using which an agent can update the amount of trust it places in other agents on an ongoing basis. This paper shows via simulation that the proposed approach (a) provides accurate estimates of the trustworthiness of agents that change behavior frequently; and (b) captures the dynamic behavior of the agents. This paper includes an evaluation based on a real dataset drawn from Amazon Marketplace, a leading e-commerce site. △ Less

Submitted 16 January, 2014; originally announced January 2014.

Journal ref: Journal Of Artificial Intelligence Research, Volume 40, pages 221-267, 2011

Tunable nonlinear PT-symmetric defect modes with an atomic cell

Authors: Chao Hang, Dmitry A. Zezyulin, Vladimir V. Konotop, Guoxiang Huang

Abstract: We propose a scheme of creating a tunable highly nonlinear defect in a one-dimensional photonic crystal. The defect consists of an atomic cell filled in with two isotopes of three-level atoms. The probe-field refractive index of the defect can be made parity-time (PT) symmetric, which is achieved by proper combination of a control field and of Stark shifts induced by a far-off-resonance field. In… ▽ More We propose a scheme of creating a tunable highly nonlinear defect in a one-dimensional photonic crystal. The defect consists of an atomic cell filled in with two isotopes of three-level atoms. The probe-field refractive index of the defect can be made parity-time (PT) symmetric, which is achieved by proper combination of a control field and of Stark shifts induced by a far-off-resonance field. In the PT-symmetric system families of stable nonlinear defect modes can be formed by the probe field. △ Less

Submitted 11 September, 2013; originally announced September 2013.

Comments: 4 pages, 2 figures, to be appeared in Opt. Lett

Parametric patterns in optical fiber ring nonlinear resonators

Authors: K. Staliunas, Chao Hang, V. V. Konotop

Abstract: We propose that parametrically excited patterns, also known as Faraday patterns, can be observed in nonlinear fiber resonators, where the coefficient of Kerr nonlinearity is periodically varying along the fiber in resonator. We study the parametric instability analytically on the basis of the Floquet theory, also numerically, by direct integration of the system. Instead of classical Faraday wave e… ▽ More We propose that parametrically excited patterns, also known as Faraday patterns, can be observed in nonlinear fiber resonators, where the coefficient of Kerr nonlinearity is periodically varying along the fiber in resonator. We study the parametric instability analytically on the basis of the Floquet theory, also numerically, by direct integration of the system. Instead of classical Faraday wave excitation scenario, where modulation in time causes formation of patterns in space, here we propose an inverted scenario, where the modulation in space excites the patterns in time. △ Less

Submitted 10 August, 2013; originally announced August 2013.

Comments: 12 pages, 5 figures, to be appeared in Phys. Rev. A

PT-symmetry with a system of three-level atoms

Authors: Chao Hang, Guoxiang Huang, Vladimir V. Konotop

Abstract: We show that a vapor of multilevel atoms driven by far-off resonant laser beams, with possibility of interference of two Raman resonances, is highly efficient for creating parity-time (PT) symmetric profiles of the probe-field refractive index, whose real part is symmetric and imaginary part is anti-symmetric in space. The spatial modulation of the susceptibility is achieved by proper combination… ▽ More We show that a vapor of multilevel atoms driven by far-off resonant laser beams, with possibility of interference of two Raman resonances, is highly efficient for creating parity-time (PT) symmetric profiles of the probe-field refractive index, whose real part is symmetric and imaginary part is anti-symmetric in space. The spatial modulation of the susceptibility is achieved by proper combination of standing-wave strong control fields and of Stark shifts induced by a far-off-resonance laser field. As particular examples we explore a mixture of isotopes of Rubidium atoms and design a PT-symmetric lattice and a parabolic refractive index with a linear imaginary part. △ Less

Submitted 21 December, 2012; originally announced December 2012.

Comments: 5 pages, 4 figures

Stern-Gerlach Effect of Weak-Light Ultraslow Vector Solitons

Authors: Chao Hang, Guoxiang Huang

Abstract: We propose a scheme to exhibit Stern-Gerlach (SG) deflection of high-dimensional vector optical soliton (VOS) at weak-light level in a cold atomic gas via electromagnetically induced transparency. We show that the propagating velocity and generation power of such VOS can be reduced to $10^{-6}\,\,c$ ($c$ is light speed in vacuum) and lowered to magnitude of nanowatt, respectively. The stabilizatio… ▽ More We propose a scheme to exhibit Stern-Gerlach (SG) deflection of high-dimensional vector optical soliton (VOS) at weak-light level in a cold atomic gas via electromagnetically induced transparency. We show that the propagating velocity and generation power of such VOS can be reduced to $10^{-6}\,\,c$ ($c$ is light speed in vacuum) and lowered to magnitude of nanowatt, respectively. The stabilization of the VOS may be realized by using an optical lattice formed by a far-detuned laser field, and its trajectory can be deflected significantly by using a SG magnetic field. Deflection angle of the VOS can be of magnitude of $10^{-3}$ rad when propagating several millimeters. Different from atomic SG deflection, deflection angle of the VOS can be distinct for different polarization components and can be manipulated in a controllable way. The results obtained can be described in terms of a SG effect for the VOS with quasispin and effective magnetic moment. △ Less

Submitted 1 September, 2012; originally announced September 2012.

Comments: 5 pages, 3 figures

Journal ref: Phys. Rev. A 86, 043809 (2012)

Highly Entangled Photons and Rapidly Responding Polarization Qubit Phase Gates in a Room-Temperature Active Raman Gain Medium

Authors: Chao Hang, Guoxiang Huang

Abstract: We present a scheme for obtaining entangled photons and quantum phase gates in a room-temperature four-state tripod-type atomic system with two-mode active Raman gain (ARG). We analyze the linear and nonlinear optical response of this ARG system and show that the scheme is fundamentally different from those based on electromagnetically induced transparency and hence can avoid significant probe-fie… ▽ More We present a scheme for obtaining entangled photons and quantum phase gates in a room-temperature four-state tripod-type atomic system with two-mode active Raman gain (ARG). We analyze the linear and nonlinear optical response of this ARG system and show that the scheme is fundamentally different from those based on electromagnetically induced transparency and hence can avoid significant probe-field absorption as well as temperature-related Doppler effect. We demonstrate that highly entangled photon pairs can be produced and rapidly responding polarization qubit phase gates can be constructed based on the unique features of enhanced cross-phase modulation and superluminal probe-field propagation of the system. △ Less

Submitted 4 December, 2011; originally announced December 2011.

Journal ref: Phys. Rev. A 82, 053818 (2010)

Linear and Nonlinear Pulse Propagations in Lifetime-Broadened Atomic Media with Spontaneously Generated Coherence

Authors: Chao Hang, Guoxiang Huang

Abstract: The linear and nonlinear pulse propagations in lifetime-broadened three-state media with spontaneously generated coherence (SGC) are investigated theoretically. Three generic systems of V-, $Λ$-, and $Ξ$-type level configurations are considered and compared. It is shown that in linear propagation regime the SGC in the V-type system can result in a significant change of dispersion and absorption an… ▽ More The linear and nonlinear pulse propagations in lifetime-broadened three-state media with spontaneously generated coherence (SGC) are investigated theoretically. Three generic systems of V-, $Λ$-, and $Ξ$-type level configurations are considered and compared. It is shown that in linear propagation regime the SGC in the V-type system can result in a significant change of dispersion and absorption and may be used to completely eliminate the absorption and largely reduce the group velocity of a probe field. However, the SGC has no effect on the dispersion and absorption of the $Λ$- and $Ξ$-type systems. In nonlinear propagation regime, the SGC displays different influences on Kerr nonlinearity for different systems. Specifically, it can enhance the Kerr nonlinearity of the V-type system whereas weaken the Kerr nonlinearity of the $Λ$-type system. Using the SGC, stable optical solitons with ultraslow propagating velocity and very low pump power can be produced in the V-type system by exploiting only one laser field in the system. △ Less

Submitted 16 November, 2011; originally announced November 2011.

Comments: 23 pages, 8 figures

Journal ref: JOSA B, Vol. 29, Issue 8, pp. 2009-2015 (2012)

Fourier mode dynamics for the nonlinear Schroedinger equation in one-dimensional bounded domains

Authors: J. G. Caputo, N. K. Efremidis, Chao Hang

Abstract: We analyze the 1D focusing nonlinear Schrödinger equation in a finite interval with homogeneous Dirichlet or Neumann boundary conditions. There are two main dynamics, the collapse which is very fast and a slow cascade of Fourier modes. For the cubic nonlinearity the calculations show no long term energy exchange between Fourier modes as opposed to higher nonlinearities. This slow dynamics is expla… ▽ More We analyze the 1D focusing nonlinear Schrödinger equation in a finite interval with homogeneous Dirichlet or Neumann boundary conditions. There are two main dynamics, the collapse which is very fast and a slow cascade of Fourier modes. For the cubic nonlinearity the calculations show no long term energy exchange between Fourier modes as opposed to higher nonlinearities. This slow dynamics is explained by fairly simple amplitude equations for the resonant Fourier modes. Their solutions are well behaved so filtering high frequencies prevents collapse. Finally these equations elucidate the unique role of the zero mode for the Neumann boundary conditions. △ Less

Submitted 31 March, 2011; originally announced March 2011.

Analytical three-dimensional bright solitons and soliton-pairs in Bose-Einstein condensates with time-space modulation

Authors: Zhenya Yan, Chao Hang

Abstract: We provide analytical three-dimensional bright multi-soliton solutions to the (3+1)-dimensional Gross-Pitaevskii (GP) equation with time and space-dependent potential, time-dependent nonlinearity, and gain/loss. The zigzag propagation trace and the breathing behavior of solitons are observed. Different shapes of bright solitons and fascinating interactions between two solitons can be achieved with… ▽ More We provide analytical three-dimensional bright multi-soliton solutions to the (3+1)-dimensional Gross-Pitaevskii (GP) equation with time and space-dependent potential, time-dependent nonlinearity, and gain/loss. The zigzag propagation trace and the breathing behavior of solitons are observed. Different shapes of bright solitons and fascinating interactions between two solitons can be achieved with different parameters. The obtained results may raise the possibility of relative experiments and potential applications. △ Less

Submitted 20 September, 2010; originally announced September 2010.

Comments: 5 pages, 4 figures

Journal ref: Phys. Rev. A, 80, 063626 (2009)

All-optical steering of light via spatial Bloch oscillations in a gas of three-level atoms

Authors: Chao Hang, V. V. Konotop

Abstract: A standing-wave control field applied to a three-level atomic medium in a planar hollow-core photonic crystal waveguide creates periodic variations of linear and nonlinear refractive indexes of the medium. This property can be used for efficient steering of light. In this work we study, both analytically and numerically, the dynamics of probe optical beams in such structures. By properly designi… ▽ More A standing-wave control field applied to a three-level atomic medium in a planar hollow-core photonic crystal waveguide creates periodic variations of linear and nonlinear refractive indexes of the medium. This property can be used for efficient steering of light. In this work we study, both analytically and numerically, the dynamics of probe optical beams in such structures. By properly designing the spatial dependence of the nonlinearity it is possible to induce long-living Bloch oscillations of spatial gap solitons, thus providing desirable change in direction of the beam propagation without inducing appreciable diffraction. Due to the significant enhancement of the nonlinearity, such self-focusing of the probe beam can be reached at extremely weak light intensities. △ Less

Submitted 19 February, 2010; originally announced February 2010.

Comments: 8 pages, 4 figures

Journal ref: Phys. Rev. A 81, 053849 (2010)

Weak-Light Ultraslow Vector Optical Solitons via Electromagnetically Induced Transparency

Authors: Chao Hang, Guoxiang Huang

Abstract: We propose a scheme to generate temporal vector optical solitons in a lifetime broadened five-state atomic medium via electromagnetically induced transparency. We show that this scheme, which is fundamentally different from the passive one by using optical fibers, is capable of achieving distortion-free vector optical solitons with ultraslow propagating velocity under very weak drive conditions.… ▽ More We propose a scheme to generate temporal vector optical solitons in a lifetime broadened five-state atomic medium via electromagnetically induced transparency. We show that this scheme, which is fundamentally different from the passive one by using optical fibers, is capable of achieving distortion-free vector optical solitons with ultraslow propagating velocity under very weak drive conditions. We demonstrate both analytically and numerically that it is easy to realize Manakov temporal vector solitons by actively manipulating the dispersion and self- and cross-phase modulation effects of the system. △ Less

Submitted 18 May, 2007; originally announced May 2007.

Comments: 4 pages, 4 figures

Three-Way Entanglement and Three-Qubit Phase Gate Based on a Coherent Six-Level Atomic System

Authors: Chao Hang, Yun Li, Lei Ma, Guoxiang Huang

Abstract: We analyze the nonlinear optical response of a six-level atomic system under a configuration of electromagnetically induced transparency. The giant fifth-order nonlinearity generated in such a system with a relatively large cross-phase modulation effect can produce efficient three-way entanglement and may be used for realizing a three-qubit quantum phase gate. We demonstrate that such phase gate… ▽ More We analyze the nonlinear optical response of a six-level atomic system under a configuration of electromagnetically induced transparency. The giant fifth-order nonlinearity generated in such a system with a relatively large cross-phase modulation effect can produce efficient three-way entanglement and may be used for realizing a three-qubit quantum phase gate. We demonstrate that such phase gate can be transferred to a Toffoli gate, facilitating practical applications in quantum information and computation. △ Less

Submitted 26 February, 2006; originally announced February 2006.

Comments: 10 pages, 2 figures

Controllable Entanglement of Lights in a Five-Level System

Authors: Yun Li, Chao Hang, Lei Ma, Guoxiang Huang

Abstract: We analyze the nonlinear optical response of a five-level system under a novel configuration of electro-magnetically induced transparency. We show that a giant Kerr nonlinearity with a relatively large cross-phase modulation coefficient that occurs in such system may be used to produce an efficient photon-photon entanglement. We demonstrate that such photon-photon entanglement is practically con… ▽ More We analyze the nonlinear optical response of a five-level system under a novel configuration of electro-magnetically induced transparency. We show that a giant Kerr nonlinearity with a relatively large cross-phase modulation coefficient that occurs in such system may be used to produce an efficient photon-photon entanglement. We demonstrate that such photon-photon entanglement is practically controllable and hence facilitates promising applications in quantum information and computation. △ Less

Submitted 16 November, 2005; v1 submitted 3 November, 2005; originally announced November 2005.

Comments: 13 pages, 4 figures, 1 column. We have added a section in which the distortion of pulses due to the dispersion is considered