arXiv subjects

Yue Zhang

Publications and source records attributed to Yue Zhang.

At least 631 records · Page 35Linked to original sources

arXiv2021

Automatic hippocampal surface generation via 3D U-net and active shape modeling with hybrid particle swarm optimization

In this paper, we proposed and validated a fully automatic pipeline for hippocampal surface generation via 3D U-net coupled with active shape modeling (ASM). Principally, the proposed pipeline consisted of three steps. In the beginning, for each magnetic resonance image, a 3D U-net was employed to obtain the automatic hippocampus segmentation at each hemisphere. Secondly, ASM was performed on a group of pre-obtained template surfaces to generate mean shape and shape variation parameters through principal component analysis. Ultimately, hybrid particle swarm optimization was utilized to search for the optimal shape variation parameters that best match the segmentation. The hippocampal surface was then generated from the mean shape and the shape variation parameters. The proposed pipeline was observed to provide hippocampal surfaces at both hemispheres with high accuracy, correct anatomical topology, and sufficient smoothness.

cs.NE↗

arXiv2021

Knowledge Enhanced Fine-Tuning for Better Handling Unseen Entities in Dialogue Generation

Although pre-training models have achieved great success in dialogue generation, their performance drops dramatically when the input contains an entity that does not appear in pre-training and fine-tuning datasets (unseen entity). To address this issue, existing methods leverage an external knowledge base to generate appropriate responses. In real-world scenario, the entity may not be included by the knowledge base or suffer from the precision of knowledge retrieval. To deal with this problem, instead of introducing knowledge base as the input, we force the model to learn a better semantic representation by predicting the information in the knowledge base, only based on the input context. Specifically, with the help of a knowledge base, we introduce two auxiliary training objectives: 1) Interpret Masked Word, which conjectures the meaning of the masked entity given the context; 2) Hypernym Generation, which predicts the hypernym of the entity based on the context. Experiment results on two dialogue corpus verify the effectiveness of our methods under both knowledge available and unavailable settings.

cs.CL↗

arXiv2021

Structural Adapters in Pretrained Language Models for AMR-to-text Generation

Pretrained language models (PLM) have recently advanced graph-to-text generation, where the input graph is linearized into a sequence and fed into the PLM to obtain its representation. However, efficiently encoding the graph structure in PLMs is challenging because such models were pretrained on natural language, and modeling structured data may lead to catastrophic forgetting of distributional knowledge. In this paper, we propose StructAdapt, an adapter method to encode graph structure into PLMs. Contrary to prior work, StructAdapt effectively models interactions among the nodes based on the graph connectivity, only training graph structure-aware adapter parameters. In this way, we incorporate task-specific knowledge while maintaining the topological structure of the graph. We empirically show the benefits of explicitly encoding graph structure into PLMs using StructAdapt, outperforming the state of the art on two AMR-to-text datasets, training only 5.1% of the PLM parameters.

cs.CL↗

arXiv2021

Smelting Gold and Silver for Improved Multilingual AMR-to-Text Generation

Recent work on multilingual AMR-to-text generation has exclusively focused on data augmentation strategies that utilize silver AMR. However, this assumes a high quality of generated AMRs, potentially limiting the transferability to the target task. In this paper, we investigate different techniques for automatically generating AMR annotations, where we aim to study which source of information yields better multilingual results. Our models trained on gold AMR with silver (machine translated) sentences outperform approaches which leverage generated silver AMR. We find that combining both complementary sources of information further improves multilingual AMR-to-text generation. Our models surpass the previous state of the art for German, Italian, Spanish, and Chinese by a large margin.

cs.CL↗

arXiv2021

General Kinetic Mixing in Gauged $U(1)_{L_μ-L_τ}$ Model for Muon $g-2$ and Dark Matter

The gauged $U(1)_{L_μ-L_τ}$ extension of the Standard Model is a very simple framework that can alleviate the tension in muon anomalous magnetic dipole moment, reinforced by the recent Fermilab measurement. We explore experimental probes of the $(g-2)_μ$ target with a general treatment of kinetic mixing between the $Z'$ gauge boson and the photon. The physical value of the kinetic mixing depends on a free parameter of the model and energy scale of a process. We find neutrino constraints on the $(g-2)_μ$ target including Borexino, CE$ν$NS, and white dwarfs are sensitive to this freedom and can be lifted if the kinetic mixing lies in proximity of zero at low momentum transfer. As a further step, we explore $L_μ-L_τ$ charged dark matter with a thermal origin and show that the same scenario of kinetic mixing can relax existing direct detection constraints and predict novel recoil energy dependence in the upcoming searches. Future joint effort of neutrino and dark matter experiments and precision spectral measurement will be the key to test such a theory.

hep-ph↗

arXiv2021

Multi-phase Liver Tumor Segmentation with Spatial Aggregation and Uncertain Region Inpainting

Multi-phase computed tomography (CT) images provide crucial complementary information for accurate liver tumor segmentation (LiTS). State-of-the-art multi-phase LiTS methods usually fused cross-phase features through phase-weighted summation or channel-attention based concatenation. However, these methods ignored the spatial (pixel-wise) relationships between different phases, hence leading to insufficient feature integration. In addition, the performance of existing methods remains subject to the uncertainty in segmentation, which is particularly acute in tumor boundary regions. In this work, we propose a novel LiTS method to adequately aggregate multi-phase information and refine uncertain region segmentation. To this end, we introduce a spatial aggregation module (SAM), which encourages per-pixel interactions between different phases, to make full use of cross-phase information. Moreover, we devise an uncertain region inpainting module (URIM) to refine uncertain pixels using neighboring discriminative features. Experiments on an in-house multi-phase CT dataset of focal liver lesions (MPCT-FLLs) demonstrate that our method achieves promising liver tumor segmentation and outperforms state-of-the-arts.

eess.IV↗

arXiv2021

Feature Curves and Surfaces of 3D Asymmetric Tensor Fields

3D asymmetric tensor fields have found many applications in science and engineering domains, such as fluid dynamics and solid mechanics. 3D asymmetric tensors can have complex eigenvalues, which makes their analysis and visualization more challenging than 3D symmetric tensors. Existing research in tensor field visualization focuses on 2D asymmetric tensor fields and 3D symmetric tensor fields. In this paper, we address the analysis and visualization of 3D asymmetric tensor fields. We introduce six topological surfaces and one topological curve, which lead to an eigenvalue space based on the tensor mode that we define. In addition, we identify several non-topological feature surfaces that are nonetheless physically important. Included in our analysis are the realizations that triple degenerate tensors are structurally stable and form curves, unlike the case for 3D symmetric tensors fields. Furthermore, there are two different ways of measuring the relative strengths of rotation and angular deformation in the tensor fields, unlike the case for 2D asymmetric tensor fields. We extract these feature surfaces using the A-patches algorithm. However, since three of our feature surfaces are quadratic, we develop a method to extract quadratic surfaces at any given accuracy. To facilitate the analysis of eigenvector fields, we visualize a hyperstreamline as a tree stem with the other two eigenvectors represented as thorns in the real domain or the dual-eigenvectors as leaves in the complex domain. To demonstrate the effectiveness of our analysis and visualization, we apply our approach to datasets from solid mechanics and fluid dynamics.

cs.GR↗

arXiv2021

Automatic Polygon Layout for Primal-Dual Visualization of Hypergraphs

N-ary relationships, which relate N entities where N is not necessarily two, can be visually represented as polygons whose vertices are the entities of the relationships. Manually generating a high-quality layout using this representation is labor-intensive. In this paper, we provide an automatic polygon layout generation algorithm for the visualization of N-ary relationships. At the core of our algorithm is a set of objective functions motivated by a number of design principles that we have identified. These objective functions are then used in an optimization framework that we develop to achieve high-quality layouts. Recognizing the duality between entities and relationships in the data, we provide a second visualization in which the roles of entities and relationships in the original data are reversed. This can lead to additional insight about the data. Furthermore, we enhance our framework for a joint optimization on the primal layout (original data) and the dual layout (where the roles of entities and relationships are reversed). This allows users to inspect their data using two complementary views. We apply our visualization approach to a number of datasets that include co-authorship data and social contact pattern data.

cs.GR↗

arXiv2021

Understanding the merging behavior patterns and evolutionary mechanism at freeway on-ramps

Understanding the merging behavior patterns at freeway on-ramps is important for assistanting the decisions of autonomous driving. This study develops a primitive-based framework to identify the driving patterns during merging processes and reveal the evolutionary mechanism at freeway on-ramps in congested traffic flow. The Nonhomogeneous Hidden Markov Model is introduced to decompose the merging processes into primitives containing semantic information. Then, the time-series K-means clustering is utilized to gather these primitives with variable-length time series into interpretable merging behavior patterns. Different from traditional state segmentation methods (e.g. Hidden Markov Model), the model proposed in this study considers the dependence of transition probability on exogenous variables, thereby revealing the influence of covariates on the evolution of driving patterns. This approach is evaluated in the merging area at a freeway on-ramp using the INTERACTION dataset. Results demonstrate that the approach provides an insight about the complicated merging processes. The findings about interpretable merging behavior patterns as well as the evolutionary mechanism can be used to design and improve the merging decision-making for autonomous vehicles.

eess.SP↗

arXiv2021

Non-Point Visible Light Transmitter Localization based on Monocular Camera

Many algorithms for visible light positioning (VLP) localization do not consider the shapes of the transmitters, which leads to the impracticality of the algorithm and the low localization accuracy. Therefore, this paper proposes a novel VLP algorithm and addresses the problems in terms of practicality and complexity by using one non-point transmitter based on a monocular. Because the shape of the transmitter is considered, the proposed algorithm is easy for practice and has wide applicability. Besides, it decreases the computation for simple geometric model and expands the coverage for there is a greater chance of receiving signals from one light than that of receiving signals from multiple lights.

eess.SP↗

arXiv2021

Neutrino As The Dark Force

We point out a novel role for the Standard Model neutrino in dark matter phenomenology where the exchange of neutrinos generates a long-range potential between dark matter particles. The resulting dark matter self interaction could be sufficiently strong to impact small-scale structure formation, without the need of any dark force carrier. This is a generic feature of theories where dark matter couples to the visible sector through the neutrino portal. It is highly testable with improved decay rate measurements at future $Z$, Higgs, and $τ$ factories, as well as precision cosmology.

hep-ph↗

arXiv2021

DialogSum: A Real-Life Scenario Dialogue Summarization Dataset

Proposal of large-scale datasets has facilitated research on deep neural models for news summarization. Deep learning can also be potentially useful for spoken dialogue summarization, which can benefit a range of real-life scenarios including customer service management and medication tracking. To this end, we propose DialogSum, a large-scale labeled dialogue summarization dataset. We conduct empirical analysis on DialogSum using state-of-the-art neural summarizers. Experimental results show unique challenges in dialogue summarization, such as spoken terms, special discourse structures, coreferences and ellipsis, pragmatics and social common sense, which require specific representation learning technologies to better deal with.

cs.CL↗

arXiv2021

On Commonsense Cues in BERT for Solving Commonsense Tasks

BERT has been used for solving commonsense tasks such as CommonsenseQA. While prior research has found that BERT does contain commonsense information to some extent, there has been work showing that pre-trained models can rely on spurious associations (e.g., data bias) rather than key cues in solving sentiment classification and other problems. We quantitatively investigate the presence of structural commonsense cues in BERT when solving commonsense tasks, and the importance of such cues for the model prediction. Using two different measures, we find that BERT does use relevant knowledge for solving the task, and the presence of commonsense knowledge is positively correlated to the model accuracy.

cs.CL↗

arXiv2021

Probing Dark Sector CP Violation with Electric Dipole Moments and Colliders

We study the experimental constraints on dark sector CP violation that enters the visible sector through a Higgs portal coupling $H^{\dagger} H Z^{\prime}_{μν} \widetilde Z^{\prime μν}$, where the $Z^{\prime}$ is from a new U(1) gauge symmetry which is assumed to couple to lepton number. We compute explicitly the leading two-loop contribution of this effective operator to the electron electric dipole moment (EDM) and show that the resulting constraints are comparable to those from direct $Z^{\prime}$ searches at electron-positron colliders when the effective operator is generated at tree level. We also examine an explicit UV completion for this effective operator that was first introduced to achieve electroweak baryogenesis and show that collider constraints from $B$-factories already exclude viable baryogenesis for $Z^{\prime}$ masses below 10 GeV, and that future electron-positron Higgs factories will exclude viable baryogenesis for $Z^{\prime}$ masses up to the $e^+e^-$ center-of-mass energy if anticipated luminosities are achieved. For higher $Z^{\prime}$ masses, the full viable baryogenesis parameter space lies within six orders of magnitude of the current upper bound on the electron EDM.

hep-ph↗

arXiv2021

Can Generative Pre-trained Language Models Serve as Knowledge Bases for Closed-book QA?

Recent work has investigated the interesting question using pre-trained language models (PLMs) as knowledge bases for answering open questions. However, existing work is limited in using small benchmarks with high test-train overlaps. We construct a new dataset of closed-book QA using SQuAD, and investigate the performance of BART. Experiments show that it is challenging for BART to remember training facts in high precision, and also challenging to answer closed-book questions even if relevant knowledge is retained. Some promising directions are found, including decoupling the knowledge memorizing process and the QA finetune process, forcing the model to recall relevant knowledge when question answering.

cs.CL↗

arXiv2021

Template-Based Named Entity Recognition Using BART

There is a recent interest in investigating few-shot NER, where the low-resource target domain has different label sets compared with a resource-rich source domain. Existing methods use a similarity-based metric. However, they cannot make full use of knowledge transfer in NER model parameters. To address the issue, we propose a template-based method for NER, treating NER as a language model ranking problem in a sequence-to-sequence framework, where original sentences and statement templates filled by candidate named entity span are regarded as the source sequence and the target sequence, respectively. For inference, the model is required to classify each candidate span based on the corresponding template scores. Our experiments demonstrate that the proposed method achieves 92.55% F1 score on the CoNLL03 (rich-resource task), and significantly better than fine-tuning BERT 10.88%, 15.34%, and 11.73% F1 score on the MIT Movie, the MIT Restaurant, and the ATIS (low-resource task), respectively.

cs.CL↗

arXiv2021

Semantic Representation for Dialogue Modeling

Although neural models have achieved competitive results in dialogue systems, they have shown limited ability in representing core semantics, such as ignoring important entities. To this end, we exploit Abstract Meaning Representation (AMR) to help dialogue modeling. Compared with the textual input, AMR explicitly provides core semantic knowledge and reduces data sparsity. We develop an algorithm to construct dialogue-level AMR graphs from sentence-level AMRs and explore two ways to incorporate AMRs into dialogue systems. Experimental results on both dialogue understanding and response generation tasks show the superiority of our model. To our knowledge, we are the first to leverage a formal semantic representation into neural dialogue modeling.

cs.CL↗

arXiv2021

G-Transformer for Document-level Machine Translation

Document-level MT models are still far from satisfactory. Existing work extend translation unit from single sentence to multiple sentences. However, study shows that when we further enlarge the translation unit to a whole document, supervised training of Transformer can fail. In this paper, we find such failure is not caused by overfitting, but by sticking around local minima during training. Our analysis shows that the increased complexity of target-to-source attention is a reason for the failure. As a solution, we propose G-Transformer, introducing locality assumption as an inductive bias into Transformer, reducing the hypothesis space of the attention from target to source. Experiments show that G-Transformer converges faster and more stably than Transformer, achieving new state-of-the-art BLEU scores for both non-pretraining and pre-training settings on three benchmark datasets.

cs.CL↗