arXiv subjects

Yating Zhang

Publications and source records attributed to Yating Zhang.

At least 19 recordsLinked to original sources

arXiv2026

On MIP Formulations for Logit-Based Multi-Purchase Choice Models and Applications

We study logit-based multi-purchase choice models and develop an exact solution methodology for the resulting assortment optimization problems, which we show are NP-hard to approximate. We introduce a hypergraph representation that captures general bundle-based choice structures and subsumes several models in the literature, including the BundleMVL-K and multivariate MNL models (Tulabandhula et al. 2023, Jasin et al. 2024). Leveraging this representation, we derive mixed-integer programming (MIP) formulations by integrating polyhedral relaxations from multilinear optimization with a perspective reformulation of the logit choice model. Our approach preserves the strength of the underlying polyhedral relaxations, yielding formulations with provably tighter linear programming (LP) bounds than the prevalent Big-M approach. We further characterize structural conditions on the hypergraph under which the formulations are locally sharp, thereby generalizing existing LP characterizations for path-based models. The framework extends naturally to heterogeneous and robust settings. Computational experiments demonstrate that the proposed formulations significantly improve both solution quality and scalability.

math.OC↗

arXiv2024

Manipulation of helical states revealed by crossed Andreev reflection

The edge states of a quantum spin Hall insulator exhibit helical properties, which has generated significant interest in the field of spintronics. Although it is predicted theoretically that Rashba spin-orbit coupling can effectively regulate edge state spin, experimental characterization of this spin control remains challenging. Here, we propose utilizing crossed Andreev reflection to probe the efficiency of gate voltage control on edge state spin. We calculate the transport properties of the quantum spin Hall isolators-superconductor heterojunction by using non-equilibrium Green's function method. We find that the probabilities of crossed Andreev reflection, electron transmission, and non-local conductance all include the relative spin rotation angle $\theta$ caused by the Rashba spin-orbit coupling in the helical edge states. Notably, when the incident energy is close to the superconducting gap, the differential conductance $G_2\propto-\cos{\theta}$. Thus, the influence of Rashba spin-orbit coupling on edge state spin can be quantitatively characterized by crossed Andreev reflection, which provides a feasible scheme for experimental testing.

cond-mat.mes-hall↗

arXiv2024

Hindered stokesian settling of discs and rods

We report measurements of the mean settling velocities for suspensions of discs and rods in the stokes regime for a number of particle aspect ratios. All these shapes display ''hindered settling'', namely, a decrease in settling speed as the solid volume fraction is increased. A comparison of our data to spheres reveals that discs and rods show less hindering than spheres at the same relative interparticle separation. The data for all six of our particle shapes may be scaled to collapse on that of spheres, with a scaling factor that depends only on the volume of the particle relative to a sphere. Despite the orientational degrees of freedom available with nonspherical particles, it thus appears that the dominant contribution to the hindered settling emerges from terms that are simply proportional to the volume of the sedimenting particles, enabling prediction of hindered settling of other nonpolar axisymmetric shapes.

cond-mat.soft↗

arXiv2024

CFCPalsy: Facial Image Synthesis with Cross-Fusion Cycle Diffusion Model for Facial Paralysis Individuals

Currently, the diagnosis of facial paralysis remains a challenging task, often relying heavily on the subjective judgment and experience of clinicians, which can introduce variability and uncertainty in the assessment process. One promising application in real-life situations is the automatic estimation of facial paralysis. However, the scarcity of facial paralysis datasets limits the development of robust machine learning models for automated diagnosis and therapeutic interventions. To this end, this study aims to synthesize a high-quality facial paralysis dataset to address this gap, enabling more accurate and efficient algorithm training. Specifically, a novel Cross-Fusion Cycle Palsy Expression Generative Model (CFCPalsy) based on the diffusion model is proposed to combine different features of facial information and enhance the visual details of facial appearance and texture in facial regions, thus creating synthetic facial images that accurately represent various degrees and types of facial paralysis. We have qualitatively and quantitatively evaluated the proposed method on the commonly used public clinical datasets of facial paralysis to demonstrate its effectiveness. Experimental results indicate that the proposed method surpasses state-of-the-art methods, generating more realistic facial images and maintaining identity consistency.

cs.CV↗

arXiv2024

Assortment Optimization Under History-Dependent Effects

This paper examines how to plan multi-period assortments when customer utility depends on historical assortments. We formulate this problem as a nonlinear integer programming model and show it is NP-hard in the presence of a negative history-dependent effect (such as a satiation effect). We build solution methodologies for obtaining global optimal solutions under a general setting that the history-dependent effects could be a mixture of positive and negative. We propose using a lifting-based framework to reformulate the problem as a mixed-integer exponential cone program that state-of-the-art solvers can solve. We also design a sequential revenue-ordered policy and show that it solves our problem to optimality in polynomial time when historical assortments positively affect customer utility (such as an addiction effect). Additionally, we identify an optimal cyclic policy for an asymptotic regime, and we also relate its length to the customer's memory length. Finally, we present a case study using a catering service dataset, showing that our model demonstrates good fitness and can effectively balance variety and revenue.

math.OC↗

arXiv2023

Precedent-Enhanced Legal Judgment Prediction with LLM and Domain-Model Collaboration

Legal Judgment Prediction (LJP) has become an increasingly crucial task in Legal AI, i.e., predicting the judgment of the case in terms of case fact description. Precedents are the previous legal cases with similar facts, which are the basis for the judgment of the subsequent case in national legal systems. Thus, it is worthwhile to explore the utilization of precedents in the LJP. Recent advances in deep learning have enabled a variety of techniques to be used to solve the LJP task. These can be broken down into two categories: large language models (LLMs) and domain-specific models. LLMs are capable of interpreting and generating complex natural language, while domain models are efficient in learning task-specific information. In this paper, we propose the precedent-enhanced LJP framework (PLJP), a system that leverages the strength of both LLM and domain models in the context of precedents. Specifically, the domain models are designed to provide candidate labels and find the proper precedents efficiently, and the large models will make the final prediction with an in-context precedents comprehension. Experiments on the real-world dataset demonstrate the effectiveness of our PLJP. Moreover, our work shows a promising direction for LLM and domain-model collaboration that can be generalized to other vertical domains.

cs.CL↗

arXiv2023

Reformulating Domain Adaptation of Large Language Models as Adapt-Retrieve-Revise: A Case Study on Chinese Legal Domain

While large language models (LLMs) like GPT-4 have recently demonstrated astonishing zero-shot capabilities in general domain tasks, they often generate content with hallucinations in specific domains such as Chinese law, hindering their application in these areas. This is typically due to the absence of training data that encompasses such a specific domain, preventing GPT-4 from acquiring in-domain knowledge. A pressing challenge is that it's not plausible to continue training LLMs of such scale on in-domain data. This paper introduces a simple and effective domain adaptation framework for GPT-4 by reformulating generation as an \textbf{adapt-retrieve-revise} process. The initial step is to \textbf{adapt} an affordable 7B LLM to the target domain by continuing learning on in-domain data. When solving a task, we leverage the adapted LLM to generate a draft answer given a task query. Then, the draft answer will be used to \textbf{retrieve} supporting evidence candidates from an external in-domain knowledge base. Finally, the draft answer and retrieved evidence are concatenated into a whole prompt to let GPT-4 assess the evidence and \textbf{revise} the draft answer to generate the final answer. Our proposal combines the advantages of the efficiency of adapting a smaller 7B model with the evidence-assessing capability of GPT-4 and effectively prevents GPT-4 from generating hallucinatory content. In the zero-shot setting of four Chinese legal tasks, our method improves accuracy by 33.3\% compared to the direct generation by GPT-4. When compared to two stronger retrieval-based baselines, our method outperforms them by 15.4\% and 23.9\%. Our code will be released

cs.CL↗

arXiv2023

SAM-Deblur: Let Segment Anything Boost Image Deblurring

Image deblurring is a critical task in the field of image restoration, aiming to eliminate blurring artifacts. However, the challenge of addressing non-uniform blurring leads to an ill-posed problem, which limits the generalization performance of existing deblurring models. To solve the problem, we propose a framework SAM-Deblur, integrating prior knowledge from the Segment Anything Model (SAM) into the deblurring task for the first time. In particular, SAM-Deblur is divided into three stages. First, we preprocess the blurred images, obtain segment masks via SAM, and propose a mask dropout method for training to enhance model robustness. Then, to fully leverage the structural priors generated by SAM, we propose a Mask Average Pooling (MAP) unit specifically designed to average SAM-generated segmented areas, serving as a plug-and-play component which can be seamlessly integrated into existing deblurring networks. Finally, we feed the fused features generated by the MAP Unit into the deblurring model to obtain a sharp image. Experimental results on the RealBlurJ, ReloBlur, and REDS datasets reveal that incorporating our methods improves GoPro-trained NAFNet's PSNR by 0.05, 0.96, and 7.03, respectively. Project page is available at GitHub \href{https://hplqaq.github.io/projects/sam-deblur}{HPLQAQ/SAM-Deblur}.

cs.CV↗

arXiv2023

PPN: Parallel Pointer-based Network for Key Information Extraction with Complex Layouts

Key Information Extraction (KIE) is a challenging multimodal task that aims to extract structured value semantic entities from visually rich documents. Although significant progress has been made, there are still two major challenges that need to be addressed. Firstly, the layout of existing datasets is relatively fixed and limited in the number of semantic entity categories, creating a significant gap between these datasets and the complex real-world scenarios. Secondly, existing methods follow a two-stage pipeline strategy, which may lead to the error propagation problem. Additionally, they are difficult to apply in situations where unseen semantic entity categories emerge. To address the first challenge, we propose a new large-scale human-annotated dataset named Complex Layout form for key information EXtraction (CLEX), which consists of 5,860 images with 1,162 semantic entity categories. To solve the second challenge, we introduce Parallel Pointer-based Network (PPN), an end-to-end model that can be applied in zero-shot and few-shot scenarios. PPN leverages the implicit clues between semantic entities to assist extracting, and its parallel extraction mechanism allows it to extract multiple results simultaneously and efficiently. Experiments on the CLEX dataset demonstrate that PPN outperforms existing state-of-the-art methods while also offering a much faster inference speed.

cs.AI↗

arXiv2023

Multidimensional Perceptron for Efficient and Explainable Long Text Classification

Because of the inevitable cost and complexity of transformer and pre-trained models, efficiency concerns are raised for long text classification. Meanwhile, in the highly sensitive domains, e.g., healthcare and legal long-text mining, potential model distrust, yet underrated and underexplored, may hatch vital apprehension. Existing methods generally segment the long text, encode each piece with the pre-trained model, and use attention or RNNs to obtain long text representation for classification. In this work, we propose a simple but effective model, Segment-aWare multIdimensional PErceptron (SWIPE), to replace attention/RNNs in the above framework. Unlike prior efforts, SWIPE can effectively learn the label of the entire text with supervised training, while perceive the labels of the segments and estimate their contributions to the long-text labeling in an unsupervised manner. As a general classifier, SWIPE can endorse different encoders, and it outperforms SOTA models in terms of classification accuracy and model efficiency. It is noteworthy that SWIPE achieves superior interpretability to transparentize long text classification results.

cs.CL↗

arXiv2022

Exploiting Contrastive Learning and Numerical Evidence for Confusing Legal Judgment Prediction

Given the fact description text of a legal case, legal judgment prediction (LJP) aims to predict the case's charge, law article and penalty term. A core problem of LJP is how to distinguish confusing legal cases, where only subtle text differences exist. Previous studies fail to distinguish different classification errors with a standard cross-entropy classification loss, and ignore the numbers in the fact description for predicting the term of penalty. To tackle these issues, in this work, first, we propose a moco-based supervised contrastive learning to learn distinguishable representations, and explore the best strategy to construct positive example pairs to benefit all three subtasks of LJP simultaneously. Second, in order to exploit the numbers in legal cases for predicting the penalty terms of certain cases, we further enhance the representation of the fact description with extracted crime amounts which are encoded by a pre-trained numeracy model. Extensive experiments on public benchmarks show that the proposed method achieves new state-of-the-art results, especially on confusing legal cases. Ablation studies also demonstrate the effectiveness of each component.

cs.CL↗

arXiv2022

Ultra-narrowband terahertz circular dichroism driven by planar metasurface supporting chiral quasi bound states in continuum

Terahertz (THz) chirality pursues customizable manipulation from narrowband to broadband. While conventional THz chirality is restricted by non-negligible linewidth and unable to handle narrowband well. Recently, the concept "quasi bound states in continuum" (quasi-BIC) is introduced to optics resonance system whose the quality factor can be extremely high with the ultra-low radiative loss, thus providing a conceptual feasibility for wave control with ultra-narrow linewidth. Herein, we construct quasi-BIC in a planar all-silicon THz metasurface with in-plane C2 and mirror symmetries breaking. Such system not only exposes the symmetry-protected BIC, but also exposes the parameter-tuned BIC assigned to single resonance type. An extremely narrow linewidth (below 0.06 GHz) with high quality factor (104 level) is obtained at quasi-BIC frequency, which achieves the ultra-narrowband THz chirality.

physics.optics↗

arXiv2021

Dialogue Inspectional Summarization with Factual Inconsistency Awareness

Dialogue summarization has been extensively studied and applied, where the prior works mainly focused on exploring superior model structures to align the input dialogue and the output summary. However, for professional dialogues (e.g., legal debate and medical diagnosis), semantic/statistical alignment can hardly fill the logical/factual gap between input dialogue discourse and summary output with external knowledge. In this paper, we mainly investigate the factual inconsistency problem for Dialogue Inspectional Summarization (DIS) under non-pretraining and pretraining settings. An innovative end-to-end dialogue summary generation framework is proposed with two auxiliary tasks: Expectant Factual Aspect Regularization (EFAR) and Missing Factual Entity Discrimination (MFED). Comprehensive experiments demonstrate that the proposed model can generate a more readable summary with accurate coverage of factual aspects as well as informing the user with potential missing facts detected from the input dialogue for further human intervention.

cs.CL↗

arXiv2021

A note on perturbation analysis for T-product based tensor singular values

In this note, we present perturbation analysis for the T-product based tensor singular values defined by Lu et al. First, the Cauchy's interlacing-type theorem for tensor singular values is given. Then, the inequalities about the difference between the singular values of two matrices proposed by Mirsky are extended to tensor cases. Finally, we introduce some useful inequalities for the singular values of tensor products and sums.

math.NA↗

arXiv2021

A Neural Conversation Generation Model via Equivalent Shared Memory Investigation

Conversation generation as a challenging task in Natural Language Generation (NLG) has been increasingly attracting attention over the last years. A number of recent works adopted sequence-to-sequence structures along with external knowledge, which successfully enhanced the quality of generated conversations. Nevertheless, few works utilized the knowledge extracted from similar conversations for utterance generation. Taking conversations in customer service and court debate domains as examples, it is evident that essential entities/phrases, as well as their associated logic and inter-relationships can be extracted and borrowed from similar conversation instances. Such information could provide useful signals for improving conversation generation. In this paper, we propose a novel reading and memory framework called Deep Reading Memory Network (DRMN) which is capable of remembering useful information of similar conversations for improving utterance generation. We apply our model to two large-scale conversation datasets of justice and e-commerce fields. Experiments prove that the proposed model outperforms the state-of-the-art approaches.

cs.CL↗

arXiv2021

Legal Judgment Prediction with Multi-Stage CaseRepresentation Learning in the Real Court Setting

Legal judgment prediction(LJP) is an essential task for legal AI. While prior methods studied on this topic in a pseudo setting by employing the judge-summarized case narrative as the input to predict the judgment, neglecting critical case life-cycle information in real court setting could threaten the case logic representation quality and prediction correctness. In this paper, we introduce a novel challenging dataset from real courtrooms to predict the legal judgment in a reasonably encyclopedic manner by leveraging the genuine input of the case -- plaintiff's claims and court debate data, from which the case's facts are automatically recognized by comprehensively understanding the multi-role dialogues of the court debate, and then learnt to discriminate the claims so as to reach the final judgment through multi-task learning. An extensive set of experiments with a large civil trial data set shows that the proposed model can more accurately characterize the interactions among claims, fact and debate for legal judgment prediction, achieving significant improvements over strong state-of-the-art baselines. Moreover, the user study conducted with real judges and law school students shows the neural predictions can also be interpretable and easily observed, and thus enhancing the trial efficiency and judgment quality.

cs.CL↗

arXiv2021

CS decomposition and GSVD for tensors based on the T-product

This paper derives the CS decomposition for orthogonal tensors (T-CSD) and the generalized singular value decomposition for two tensors (T-GSVD) via the T-product. The structures of the two decompositions are analyzed in detail and are consistent with those for matrix cases. Then the corresponding algorithms are proposed respectively. Finally, T-GSVD can be used to give the explicit expression for the solution of tensor Tikhonov regularization. Numerical examples demonstrate the effectiveness of T-GSVD in solving image restoration problems.

math.NA↗

arXiv2020

Cross Copy Network for Dialogue Generation

In the past few years, audiences from different fields witness the achievements of sequence-to-sequence models (e.g., LSTM+attention, Pointer Generator Networks, and Transformer) to enhance dialogue content generation. While content fluency and accuracy often serve as the major indicators for model training, dialogue logics, carrying critical information for some particular domains, are often ignored. Take customer service and court debate dialogue as examples, compatible logics can be observed across different dialogue instances, and this information can provide vital evidence for utterance generation. In this paper, we propose a novel network architecture - Cross Copy Networks(CCN) to explore the current dialog context and similar dialogue instances' logical structure simultaneously. Experiments with two tasks, court debate and customer service content generation, proved that the proposed algorithm is superior to existing state-of-art content generation models.

cs.CL↗