arXiv subjects

Yitong Li

Publications and source records attributed to Yitong Li.

At least 91 records · Page 5Linked to original sources

arXiv2023

Scribble-Supervised Target Extraction Method Based on Inner Structure-Constraint for Remote Sensing Images

Weakly supervised learning based on scribble annotations in target extraction of remote sensing images has drawn much interest due to scribbles' flexibility in denoting winding objects and low cost of manually labeling. However, scribbles are too sparse to identify object structure and detailed information, bringing great challenges in target localization and boundary description. To alleviate these problems, in this paper, we construct two inner structure-constraints, a deformation consistency loss and a trainable active contour loss, together with a scribble-constraint to supervise the optimization of the encoder-decoder network without introducing any auxiliary module or extra operation based on prior cues. Comprehensive experiments demonstrate our method's superiority over five state-of-the-art algorithms in this field. Source code is available at https://github.com/yitongli123/ISC-TE.

cs.CV↗

arXiv2023

CHBias: Bias Evaluation and Mitigation of Chinese Conversational Language Models

\textit{\textbf{\textcolor{red}{Warning}:} This paper contains content that may be offensive or upsetting.} Pretrained conversational agents have been exposed to safety issues, exhibiting a range of stereotypical human biases such as gender bias. However, there are still limited bias categories in current research, and most of them only focus on English. In this paper, we introduce a new Chinese dataset, CHBias, for bias evaluation and mitigation of Chinese conversational language models. Apart from those previous well-explored bias categories, CHBias includes under-explored bias categories, such as ageism and appearance biases, which received less attention. We evaluate two popular pretrained Chinese conversational models, CDial-GPT and EVA2.0, using CHBias. Furthermore, to mitigate different biases, we apply several debiasing methods to the Chinese pretrained models. Experimental results show that these Chinese pretrained models are potentially risky for generating texts that contain social biases, and debiasing methods using the proposed dataset can make response generation less biased while preserving the models' conversational capabilities.

cs.CL↗

arXiv2023

On the Importance of Accurate Geometry Data for Dense 3D Vision Tasks

Learning-based methods to solve dense 3D vision problems typically train on 3D sensor data. The respectively used principle of measuring distances provides advantages and drawbacks. These are typically not compared nor discussed in the literature due to a lack of multi-modal datasets. Texture-less regions are problematic for structure from motion and stereo, reflective material poses issues for active sensing, and distances for translucent objects are intricate to measure with existing hardware. Training on inaccurate or corrupt data induces model bias and hampers generalisation capabilities. These effects remain unnoticed if the sensor measurement is considered as ground truth during the evaluation. This paper investigates the effect of sensor errors for the dense 3D vision tasks of depth estimation and reconstruction. We rigorously show the significant impact of sensor characteristics on the learned predictions and notice generalisation issues arising from various technologies in everyday household environments. For evaluation, we introduce a carefully designed dataset\footnote{dataset available at https://github.com/Junggy/HAMMER-dataset} comprising measurements from commodity sensors, namely D-ToF, I-ToF, passive/active stereo, and monocular RGB+P. Our study quantifies the considerable sensor noise impact and paves the way to improved dense vision estimates and targeted data fusion.

cs.CV↗

arXiv2023

The Intrinsic Communication in Power Systems: A New Perspective to Understand Synchronization Stability

Synchronization is an essential element in three-phase ac electric power systems. The large-scale integration of converter-interfaced resources leads to the power grid transformation from voltage-source-dominated to voltage-current-source-composite, which also raises new challenges to model and analyze the system synchronization. In this article, we present the intrinsic analogy of a power system to a communication system, which is here called power-communication isomorphism. Based on this isomorphism, we revisit power system synchronization stability from a communication perspective and thereby establish a theory that unifies the synchronization dynamics of heterogeneous power apparatuses. In particular, the proposed theory is used to interpret and analyze, for example, the different power control speeds of power apparatuses; the role of network line dynamics in synchronization analysis; and the synchronization capability of current-source-dominated (e.g., grid-following-converter-dominated) grids. The findings are verified on the IEEE 68-bus test system.

eess.SY↗

arXiv2023

Whole-System First-Swing Stability of Inverter-Based Inertia-Free Power Systems

The emphasis on inertia for system stability has been a long-held tradition in conventional grids. The fast and flexible controllability of inverters opens up new possibilities. This paper investigates the {first-swing stability} of inverter-based inertia-free power systems. We illustrate that by replacing inertia with fast primary control (i.e. inertia-free), the first-swing stability region is greatly extended compared to the classic equal area criterion of inertia-rich systems. The extended stability region is fully decentralised and ensures whole-system stability for all swing modes across all timescales and interaction boundaries in a complex grid. The removal of inertia has impacts on frequency stability but these impacts can be well mitigated. The findings of the paper are proved mathematically and verified by simulation on the IEEE 68-bus system.

eess.SY↗

arXiv2023

Descriptor State Space Modeling of Power Systems

State space is widely used for modeling power systems and analyzing their dynamics but it is limited to representing causal and proper systems in which the number of zeros does not exceed the number of poles. In other words, the system input, output, and state can not be freely selected. This limits how flexibly models are constructed, and in some circumstances, can introduce errors because of the addition of virtual elements in order to connect the mismatched ports of subsystem models. An extension known as descriptor state space can model both proper and improper systems and is a promising candidate for solving the noted problems. It facilitates a modular construction of power system models with flexible choice of ports of subsystems. Algorithms for mathematical manipulation of descriptor state space models are derived such as preforming inverse, connection, and transform. Corresponding physical interpretations are also given. Importantly, the proposed algorithms preserve the subsystem states in the whole system model, which therefore enables the analysis of root causes of instability and mode participation. Theoretical advances are validated by example power systems of varied scales including a single-inductor or -capacitor system, two-inductor or -capacitor system, and a modified IEEE 14-bus generator-inverter-composite system.

eess.SY↗

arXiv2022

Observation of Acoustic Non-Hermitian Bloch Braids and Associated Topological Phase Transitions

Topological features embedded in ancient braiding and knotting arts endow significant impacts on our daily life and even cutting-edge science. Recently, fast growing efforts are invested to the braiding topology of complex Bloch bands in non-Hermitian systems. This new classification of band topology goes far beyond those established in Hermitian counterparts. Here, we present the first acoustic realization of the topological non-Hermitian Bloch braids, based on a two-band model easily accessible for realizing any desired knot structure. The non-Hermitian bands are synthesized by a simple binary cavity-tube system, where the long-range, complex-valued, and momentum-resolved couplings are accomplished by a well-controlled unidirectional coupler. In addition to directly visualizing various two-band braiding patterns, we unambiguously observe the highly-elusive topological phase transitions between them. Not only do our results provide a direct demonstration for the non-Hermitian band topology, but also the experimental techniques open new avenues for designing unconventional acoustic metamaterials.

cond-mat.mes-hall↗

arXiv2022

PiPs: a Kernel-based Optimization Scheme for Analyzing Non-Stationary 1D Signals

This paper proposes a novel kernel-based optimization scheme to handle tasks in the analysis, e.g., signal spectral estimation and single-channel source separation of 1D non-stationary oscillatory data. The key insight of our optimization scheme for reconstructing the time-frequency information is that when a nonparametric regression is applied on some input values, the output regressed points would lie near the oscillatory pattern of the oscillatory 1D signal only if these input values are a good approximation of the ground-truth phase function. In this work, Gaussian Process (GP) is chosen to conduct this nonparametric regression: the oscillatory pattern is encoded as the Pattern-inducing Points (PiPs) which act as the training data points in the GP regression; while the targeted phase function is fed in to compute the correlation kernels, acting as the testing input. Better approximated phase function generates more precise kernels, thus resulting in smaller optimization loss error when comparing the kernel-based regression output with the original signals. To the best of our knowledge, this is the first algorithm that can satisfactorily handle fully non-stationary oscillatory data, close and crossover frequencies, and general oscillatory patterns. Even in the example of a signal {produced by slow variation in the parameters of a trigonometric expansion}, we show that PiPs admits competitive or better performance in terms of accuracy and robustness than existing state-of-the-art algorithms.

stat.ML↗

arXiv2022

Automated Integration of Infrastructure Component Status for Real-Time Restoration Progress Control: Case Study of Highway System in Hurricane Harvey

Following extreme events, efficient restoration of infrastructure systems is critical to sustaining community lifelines. During the process, effective monitoring and control of the infrastructure restoration progress is critical. This research proposes a systematic approach that automatically integrates component-level restoration status to achieve real-time forecasting of overall infrastructure restoration progress. In this research, the approach is mainly designed for transportation infrastructure restoration following Hurricane Harvey. In detail, the component-level restoration status is linked to the restoration progress forecasting through network modeling and earned value method. Once the new component restoration status is collected, the information is automatically integrated to update the overall restoration progress forecasting. Academically, an approach is proposed to automatically transform the component-level restoration information to overall restoration progress. In practice, the approach expects to ease the communication and coordination efforts between emergency managers, thereby facilitating timely identification and resolution of issues for rapid infrastructure restoration.

cs.CE↗

arXiv2022

Discrete Event Simulation for Port Berth Maintenance Planning

Industrial and commercial ports, which are one of the three main hubs to the country, require 24/7 operations to maintain the goods export and import flow. Due to the aging and weather factors, berths require regular maintenance, such as replacing old piles, timber finders, marine ladders, rubber fenders, and deck slabs. For efficient berth maintenance, strategies are highly desired to minimize or eliminate any delays in operations during the maintenance. This paper develops a discrete event simulation model using Simphony.NET for berth maintenance processes in Doha Port, Kuwait. The model derives minimum maintenance duration under limited resources and associated uncertainties. The model can be used as a decision support tool to minimize interruption or delays in the port maintenance operations.

cs.CE↗

arXiv2022

KPT: Keyword-guided Pre-training for Grounded Dialog Generation

Incorporating external knowledge into the response generation process is essential to building more helpful and reliable dialog agents. However, collecting knowledge-grounded conversations is often costly, calling for a better pre-trained model for grounded dialog generation that generalizes well w.r.t. different types of knowledge. In this work, we propose KPT (Keyword-guided Pre-Training), a novel self-supervised pre-training method for grounded dialog generation without relying on extra knowledge annotation. Specifically, we use a pre-trained language model to extract the most uncertain tokens in the dialog as keywords. With these keywords, we construct two kinds of knowledge and pre-train a knowledge-grounded response generation model, aiming at handling two different scenarios: (1) the knowledge should be faithfully grounded; (2) it can be selectively used. For the former, the grounding knowledge consists of keywords extracted from the response. For the latter, the grounding knowledge is additionally augmented with keywords extracted from other utterances in the same dialog. Since the knowledge is extracted from the dialog itself, KPT can be easily performed on a large volume and variety of dialogue data. We considered three data sources (open-domain, task-oriented, conversational QA) with a total of 2.5M dialogues. We conduct extensive experiments on various few-shot knowledge-grounded generation tasks, including grounding on dialog acts, knowledge graphs, persona descriptions, and Wikipedia passages. Our comprehensive experiments and analyses demonstrate that KPT consistently outperforms state-of-the-art methods on these tasks with diverse grounding knowledge.

cs.CL↗

arXiv2022

Towards Diverse, Relevant and Coherent Open-Domain Dialogue Generation via Hybrid Latent Variables

Conditional variational models, using either continuous or discrete latent variables, are powerful for open-domain dialogue response generation. However, previous works show that continuous latent variables tend to reduce the coherence of generated responses. In this paper, we also found that discrete latent variables have difficulty capturing more diverse expressions. To tackle these problems, we combine the merits of both continuous and discrete latent variables and propose a Hybrid Latent Variable (HLV) method. Specifically, HLV constrains the global semantics of responses through discrete latent variables and enriches responses with continuous latent variables. Thus, we diversify the generated responses while maintaining relevance and coherence. In addition, we propose Conditional Hybrid Variational Transformer (CHVT) to construct and to utilize HLV with transformers for dialogue generation. Through fine-grained symbolic-level semantic information and additive Gaussian mixing, we construct the distribution of continuous variables, prompting the generation of diverse expressions. Meanwhile, to maintain the relevance and coherence, the discrete latent variable is optimized by self-separation training. Experimental results on two dialogue generation datasets (DailyDialog and Opensubtitles) show that CHVT is superior to traditional transformer-based variational mechanism w.r.t. diversity, relevance and coherence metrics. Moreover, we also demonstrate the benefit of applying HLV to fine-tuning two pre-trained dialogue models (PLATO and BART-base).

cs.CL↗

arXiv2022

Modeling Complex Dialogue Mappings via Sentence Semantic Segmentation Guided Conditional Variational Auto-Encoder

Complex dialogue mappings (CDM), including one-to-many and many-to-one mappings, tend to make dialogue models generate incoherent or dull responses, and modeling these mappings remains a huge challenge for neural dialogue systems. To alleviate these problems, methods like introducing external information, reconstructing the optimization function, and manipulating data samples are proposed, while they primarily focus on avoiding training with CDM, inevitably weakening the model's ability of understanding CDM in human conversations and limiting further improvements in model performance. This paper proposes a Sentence Semantic \textbf{Seg}mentation guided \textbf{C}onditional \textbf{V}ariational \textbf{A}uto-\textbf{E}ncoder (SegCVAE) method which can model and take advantages of the CDM data. Specifically, to tackle the incoherent problem caused by one-to-many, SegCVAE uses response-related prominent semantics to constrained the latent variable. To mitigate the non-diverse problem brought by many-to-one, SegCVAE segments multiple prominent semantics to enrich the latent variables. Three novel components, Internal Separation, External Guidance, and Semantic Norms, are proposed to achieve SegCVAE. On dialogue generation tasks, both the automatic and human evaluation results show that SegCVAE achieves new state-of-the-art performance.

cs.CL↗

arXiv2022

An Overview on Controllable Text Generation via Variational Auto-Encoders

Recent advances in neural-based generative modeling have reignited the hopes of having computer systems capable of conversing with humans and able to understand natural language. The employment of deep neural architectures has been largely explored in a multitude of context and tasks to fulfill various user needs. On one hand, producing textual content that meets specific requirements is of priority for a model to seamlessly conduct conversations with different groups of people. On the other hand, latent variable models (LVM) such as variational auto-encoders (VAEs) as one of the most popular genres of generative models are designed to characterize the distributional pattern of textual data. Thus they are inherently capable of learning the integral textual features that are worth exploring for controllable pursuits. \noindent This overview gives an introduction to existing generation schemes, problems associated with text variational auto-encoders, and a review of several applications about the controllable generation that are instantiations of these general formulations,\footnote{A detailed paper list is available at \url{https://github.com/ImKeTT/CTG-latentAEs}} as well as related datasets, metrics and discussions for future researches. Hopefully, this overview will provide an overview of living questions, popular methodologies and raw thoughts for controllable language generation under the scope of variational auto-encoder.

cs.CL↗

arXiv2022

Towards Identifying Social Bias in Dialog Systems: Frame, Datasets, and Benchmarks

The research of open-domain dialog systems has been greatly prospered by neural models trained on large-scale corpora, however, such corpora often introduce various safety problems (e.g., offensive languages, biases, and toxic behaviors) that significantly hinder the deployment of dialog systems in practice. Among all these unsafe issues, addressing social bias is more complex as its negative impact on marginalized populations is usually expressed implicitly, thus requiring normative reasoning and rigorous analysis. In this paper, we focus our investigation on social bias detection of dialog safety problems. We first propose a novel Dial-Bias Frame for analyzing the social bias in conversations pragmatically, which considers more comprehensive bias-related analyses rather than simple dichotomy annotations. Based on the proposed framework, we further introduce CDail-Bias Dataset that, to our knowledge, is the first well-annotated Chinese social bias dialog dataset. In addition, we establish several dialog bias detection benchmarks at different label granularities and input types (utterance-level and context-level). We show that the proposed in-depth analyses together with these benchmarks in our Dial-Bias Frame are necessary and essential to bias detection tasks and can benefit building safe dialog systems in practice.

cs.CL↗

arXiv2022

A multi-band atomic candle with microwave-dressed Rydberg atoms

Stabilizing important physical quantities to atom-based standards lies at the heart of modern atomic, molecular and optical physics, and is widely applied to the field of precision metrology. Of particular importance is the atom-based microwave field amplitude stabilizer, the so-called atomic candle. Previous atomic candles are realized with atoms in their ground state, and hence suffer from the lack of frequency band tunability and small stabilization bandwidth, severely limiting their development and potential applications. To tackle these limitations, we employ microwave-dressed Rydberg atoms to realize a novel atomic candle that features multi-band frequency tunability and large stabilization bandwidth. We demonstrate amplitude stabilization of microwave field from C-band to Ka-band, which could be extended to quasi-DC and terahertz fields by exploring abundant Rydberg levels. Our atomic candle achieves stabilization bandwidth of 100 Hz, outperforming previous ones by more than two orders of magnitude. Our simulation indicates the stabilization bandwidth can be further increased up to 100 kHz. Our work paves a route to develop novel electric field control and applications with a noise-resilient, miniaturized, sensitive and broadband atomic candle.

physics.atom-ph↗

arXiv2022

Pan More Gold from the Sand: Refining Open-domain Dialogue Training with Noisy Self-Retrieval Generation

Real human conversation data are complicated, heterogeneous, and noisy, from which building open-domain dialogue systems remains a challenging task. In fact, such dialogue data still contains a wealth of information and knowledge, however, they are not fully explored. In this paper, we show existing open-domain dialogue generation methods that memorize context-response paired data with autoregressive or encode-decode language models underutilize the training data. Different from current approaches, using external knowledge, we explore a retrieval-generation training framework that can take advantage of the heterogeneous and noisy training data by considering them as "evidence". In particular, we use BERTScore for retrieval, which gives better qualities of the evidence and generation. Experiments over publicly available datasets demonstrate that our method can help models generate better responses, even such training data are usually impressed as low-quality data. Such performance gain is comparable with those improved by enlarging the training set, even better. We also found that the model performance has a positive correlation with the relevance of the retrieved evidence. Moreover, our method performed well on zero-shot experiments, which indicates that our method can be more robust to real-world data.

cs.CL↗

arXiv2022

CheXplaining in Style: Counterfactual Explanations for Chest X-rays using StyleGAN

Deep learning models used in medical image analysis are prone to raising reliability concerns due to their black-box nature. To shed light on these black-box models, previous works predominantly focus on identifying the contribution of input features to the diagnosis, i.e., feature attribution. In this work, we explore counterfactual explanations to identify what patterns the models rely on for diagnosis. Specifically, we investigate the effect of changing features within chest X-rays on the classifier's output to understand its decision mechanism. We leverage a StyleGAN-based approach (StyleEx) to create counterfactual explanations for chest X-rays by manipulating specific latent directions in their latent space. In addition, we propose EigenFind to significantly reduce the computation time of generated explanations. We clinically evaluate the relevancy of our counterfactual explanations with the help of radiologists. Our code is publicly available.

eess.IV↗