Searcharxiv⌕ Search

arXiv subjects

Xiao Zhang

Publications and source records attributed to Xiao Zhang.

At least 343 records · Page 19Linked to original sources
arXiv2023

SciMRC: Multi-perspective Scientific Machine Reading Comprehension

Scientific machine reading comprehension (SMRC) aims to understand scientific texts through interactions with humans by given questions. As far as we know, there is only one dataset focused on exploring full-text scientific machine reading comprehension. However, the dataset has ignored the fact that different readers may have different levels of understanding of the text, and only includes single-perspective question-answer pairs, leading to a lack of consideration of different perspectives. To tackle the above problem, we propose a novel multi-perspective SMRC dataset, called SciMRC, which includes perspectives from beginners, students and experts. Our proposed SciMRC is constructed from 741 scientific papers and 6,057 question-answer pairs. Each perspective of beginners, students and experts contains 3,306, 1,800 and 951 QA pairs, respectively. The extensive experiments on SciMRC by utilizing pre-trained models suggest the importance of considering perspectives of SMRC, and demonstrate its challenging nature for machine comprehension.

cs.CL↗
arXiv2023

Safe Reinforcement Learning with Dead-Ends Avoidance and Recovery

Safety is one of the main challenges in applying reinforcement learning to realistic environmental tasks. To ensure safety during and after training process, existing methods tend to adopt overly conservative policy to avoid unsafe situations. However, overly conservative policy severely hinders the exploration, and makes the algorithms substantially less rewarding. In this paper, we propose a method to construct a boundary that discriminates safe and unsafe states. The boundary we construct is equivalent to distinguishing dead-end states, indicating the maximum extent to which safe exploration is guaranteed, and thus has minimum limitation on exploration. Similar to Recovery Reinforcement Learning, we utilize a decoupled RL framework to learn two policies, (1) a task policy that only considers improving the task performance, and (2) a recovery policy that maximizes safety. The recovery policy and a corresponding safety critic are pretrained on an offline dataset, in which the safety critic evaluates upper bound of safety in each state as awareness of environmental safety for the agent. During online training, a behavior correction mechanism is adopted, ensuring the agent to interact with the environment using safe actions only. Finally, experiments of continuous control tasks demonstrate that our approach has better task performance with less safety violations than state-of-the-art algorithms.

cs.LG↗
arXiv2023

Discourse Representation Structure Parsing for Chinese

Previous work has predominantly focused on monolingual English semantic parsing. We, instead, explore the feasibility of Chinese semantic parsing in the absence of labeled data for Chinese meaning representations. We describe the pipeline of automatically collecting the linearized Chinese meaning representation data for sequential-to sequential neural networks. We further propose a test suite designed explicitly for Chinese semantic parsing, which provides fine-grained evaluation for parsing performance, where we aim to study Chinese parsing difficulties. Our experimental results show that the difficulty of Chinese semantic parsing is mainly caused by adverbs. Realizing Chinese parsing through machine translation and an English parser yields slightly lower performance than training a model directly on Chinese data.

cs.CL↗
arXiv2023

Effect of dynamical screening in the Bethe-Salpeter framework: Excitons in crystalline naphthalene

Solving the Bethe-Salpeter equation (BSE) for the optical polarization functions is a first principles means to model optical properties of materials including excitonic effects. One almost ubiquitously used approximation neglects the frequency dependence of the screened electron-hole interaction. This is commonly justified by the large difference in magnitude of electronic plasma frequency and exciton binding energy. We incorporated dynamical effects into the screening of the electron-hole interaction in the BSE using two different approximations as well as exact diagonalization of the exciton Hamiltonian. We compare these approaches for a naphthalene organic crystal, for which the difference between exciton binding energy and plasma frequency is only about a factor of ten. Our results show that in this case, corrections due to dynamical screening are about 15\,\% of the exciton binding energy. We analyze the effect of screening dynamics on optical absorption across the visible spectral range and use our data to establish an \emph{effective} screening model as a computationally efficient approach to approximate dynamical effects in complex materials in the future.

cond-mat.mtrl-sci↗
arXiv2023

PIXIU: A Large Language Model, Instruction Data and Evaluation Benchmark for Finance

Although large language models (LLMs) has shown great performance on natural language processing (NLP) in the financial domain, there are no publicly available financial tailtored LLMs, instruction tuning datasets, and evaluation benchmarks, which is critical for continually pushing forward the open-source development of financial artificial intelligence (AI). This paper introduces PIXIU, a comprehensive framework including the first financial LLM based on fine-tuning LLaMA with instruction data, the first instruction data with 136K data samples to support the fine-tuning, and an evaluation benchmark with 5 tasks and 9 datasets. We first construct the large-scale multi-task instruction data considering a variety of financial tasks, financial document types, and financial data modalities. We then propose a financial LLM called FinMA by fine-tuning LLaMA with the constructed dataset to be able to follow instructions for various financial tasks. To support the evaluation of financial LLMs, we propose a standardized benchmark that covers a set of critical financial tasks, including five financial NLP tasks and one financial prediction task. With this benchmark, we conduct a detailed analysis of FinMA and several existing LLMs, uncovering their strengths and weaknesses in handling critical financial tasks. The model, datasets, benchmark, and experimental results are open-sourced to facilitate future research in financial AI.

cs.CL↗
arXiv2023

Evolution of magnetic correlation in an inhomogeneous square lattice

We explore the magnetic properties of a two-dimensional Hubbard model on an inhomogeneous square lattice, which provides a platform for tuning the bandwidth of the flat band. In its limit, this inhomogeneous square lattice turns into a Lieb lattice, and it exhibits abundant properties due to the flat band structure at the Fermi level. By using the determinant quantum Monte Carlo simulation, we calculate the spin susceptibility, double occupancy, magnetization, spin structure factor, and effective pairing interaction of the system. It is found that the antiferromagnetic correlation is suppressed by the inhomogeneous strength and that the ferromagnetic correlation is enhanced. Both the antiferromagnetic correlation and ferromagnetic correlation are enhanced as the interaction increases. It is also found that the effective $d$-wave pairing interaction is suppressed by the increasing inhomogeneity. In addition, we also study the thermodynamic properties of the inhomogeneous square lattice, and the calculation of specific heat provide good support for our point. Our intensive numerical results provide a rich magnetic phase diagram over both the inhomogeneity and interaction.

cond-mat.str-el↗
arXiv2023

Quantum Monte Carlo study of superconductivity in rhombohedral trilayer graphene under an electric field

By using the constrained-phase quantum Monte Carlo method, we performed a systematic study of the ground state of the half filled Hubbard model for a trilayer honeycomb lattice. We analyze the effect of the perpendicular electric field on the electronic structure, magnetic property and pairing correlations. It is found that the antiferromagnetism is suppressed by the perpendicular electric field, especially the long-range parts, and the dominant magnetic fluctuations are still antiferromagnetic. The electronic correlation drives a $d+id$ superconducting pairing to be dominant over other pairing patterns among various electric fields and interaction strengths. We also found that the $d+id$ pairing correlation is greatly enhanced as the on-site Coulomb interaction is increased. Our intensive numerical results may unveil the nature of the recently observed superconductivity in rhombohedral trilayer graphene under an electric field.

cond-mat.str-el↗
arXiv2023

The Dirac equation on metrics of Eguchi-Hanson type

We investigate parallel spinors on the Eguchi-Hanson metrics and find the space of complex parallel spinors are complex 2-dimensional. For the metrics of Eguchi-Hanson type with the zero scalar curvature, we separate variables for the harmonic spinors and obtain the solutions explicitly.

math.DG↗
arXiv2023

Geodesics on metrics of Eguchi-Hanson type

Geodesic equations are solved when at least two of $θ$, $ϕ$ and $ψ$ are constant, or $r$ is constant, on scalar flat metrics of Eguchi-Hanson type. They can also be solved also on Eguchi-Hanson metrics which are Ricci flat if only $ϕ$ is constant. However, the explicit solution of the geodesic equations is not available yet if only $ψ$ is constant.

gr-qc↗
arXiv2023

Robust Image Ordinal Regression with Controllable Image Generation

Image ordinal regression has been mainly studied along the line of exploiting the order of categories. However, the issues of class imbalance and category overlap that are very common in ordinal regression were largely overlooked. As a result, the performance on minority categories is often unsatisfactory. In this paper, we propose a novel framework called CIG based on controllable image generation to directly tackle these two issues. Our main idea is to generate extra training samples with specific labels near category boundaries, and the sample generation is biased toward the less-represented categories. To achieve controllable image generation, we seek to separate structural and categorical information of images based on structural similarity, categorical similarity, and reconstruction constraints. We evaluate the effectiveness of our new CIG approach in three different image ordinal regression scenarios. The results demonstrate that CIG can be flexibly integrated with off-the-shelf image encoders or ordinal regression models to achieve improvement, and further, the improvement is more significant for minority categories.

cs.CV↗
arXiv2023

Contracted Product of Hypermatrices via STP of Matrices

An equivalent definition of hypermatrices is introduced. The matrix expression of hypermatrices is proposed. Using permutation matrices, the conversion of different matrix expressions is revealed. The various contracted products of hypermatrices are realized by semi-tensor products (STP) of matrices via matrix expressions of hypermatrices.

math.NA↗
arXiv2023

When Search Meets Recommendation: Learning Disentangled Search Representation for Recommendation

Modern online service providers such as online shopping platforms often provide both search and recommendation (S&R) services to meet different user needs. Rarely has there been any effective means of incorporating user behavior data from both S&R services. Most existing approaches either simply treat S&R behaviors separately, or jointly optimize them by aggregating data from both services, ignoring the fact that user intents in S&R can be distinctively different. In our paper, we propose a Search-Enhanced framework for the Sequential Recommendation (SESRec) that leverages users' search interests for recommendation, by disentangling similar and dissimilar representations within S&R behaviors. Specifically, SESRec first aligns query and item embeddings based on users' query-item interactions for the computations of their similarities. Two transformer encoders are used to learn the contextual representations of S&R behaviors independently. Then a contrastive learning task is designed to supervise the disentanglement of similar and dissimilar representations from behavior sequences of S&R. Finally, we extract user interests by the attention mechanism from three perspectives, i.e., the contextual representations, the two separated behaviors containing similar and dissimilar interests. Extensive experiments on both industrial and public datasets demonstrate that SESRec consistently outperforms state-of-the-art models. Empirical studies further validate that SESRec successfully disentangle similar and dissimilar user interests from their S&R behaviors.

cs.IR↗
arXiv2023

Law Article-Enhanced Legal Case Matching: a Causal Learning Approach

Legal case matching, which automatically constructs a model to estimate the similarities between the source and target cases, has played an essential role in intelligent legal systems. Semantic text matching models have been applied to the task where the source and target legal cases are considered as long-form text documents. These general-purpose matching models make the predictions solely based on the texts in the legal cases, overlooking the essential role of the law articles in legal case matching. In the real world, the matching results (e.g., relevance labels) are dramatically affected by the law articles because the contents and the judgments of a legal case are radically formed on the basis of law. From the causal sense, a matching decision is affected by the mediation effect from the cited law articles by the legal cases, and the direct effect of the key circumstances (e.g., detailed fact descriptions) in the legal cases. In light of the observation, this paper proposes a model-agnostic causal learning framework called Law-Match, under which the legal case matching models are learned by respecting the corresponding law articles. Given a pair of legal cases and the related law articles, Law-Match considers the embeddings of the law articles as instrumental variables (IVs), and the embeddings of legal cases as treatments. Using IV regression, the treatments can be decomposed into law-related and law-unrelated parts, respectively reflecting the mediation and direct effects. These two parts are then combined with different weights to collectively support the final matching prediction. We show that the framework is model-agnostic, and a number of legal case matching models can be applied as the underlying models. Comprehensive experiments show that Law-Match can outperform state-of-the-art baselines on three public datasets.

cs.IR↗
arXiv2023

A Lightweight Recurrent Learning Network for Sustainable Compressed Sensing

Recently, deep learning-based compressed sensing (CS) has achieved great success in reducing the sampling and computational cost of sensing systems and improving the reconstruction quality. These approaches, however, largely overlook the issue of the computational cost; they rely on complex structures and task-specific operator designs, resulting in extensive storage and high energy consumption in CS imaging systems. In this paper, we propose a lightweight but effective deep neural network based on recurrent learning to achieve a sustainable CS system; it requires a smaller number of parameters but obtains high-quality reconstructions. Specifically, our proposed network consists of an initial reconstruction sub-network and a residual reconstruction sub-network. While the initial reconstruction sub-network has a hierarchical structure to progressively recover the image, reducing the number of parameters, the residual reconstruction sub-network facilitates recurrent residual feature extraction via recurrent learning to perform both feature fusion and deep reconstructions across different scales. In addition, we also demonstrate that, after the initial reconstruction, feature maps with reduced sizes are sufficient to recover the residual information, and thus we achieved a significant reduction in the amount of memory required. Extensive experiments illustrate that our proposed model can achieve a better reconstruction quality than existing state-of-the-art CS algorithms, and it also has a smaller number of network parameters than these algorithms. Our source codes are available at: https://github.com/C66YU/CSRN.

cs.CV↗
arXiv2023

Transition System Representation of Boolean Control Networks

First, the topological structure of a transition system is studied. Then, two types of transition system (TS) representations of Boolean networks (BNs) and Boolean control networks (BCNs) are investigated. The first kind of representation is state-based, which converts a BCN into a TS with either distinct control or non-distinct control. The second representation is output-based, which is also called the simulation of the original BCN. Some applications are also studied.

eess.SY↗
arXiv2023

CLCLSA: Cross-omics Linked embedding with Contrastive Learning and Self Attention for multi-omics integration with incomplete multi-omics data

Integration of heterogeneous and high-dimensional multi-omics data is becoming increasingly important in understanding genetic data. Each omics technique only provides a limited view of the underlying biological process and integrating heterogeneous omics layers simultaneously would lead to a more comprehensive and detailed understanding of diseases and phenotypes. However, one obstacle faced when performing multi-omics data integration is the existence of unpaired multi-omics data due to instrument sensitivity and cost. Studies may fail if certain aspects of the subjects are missing or incomplete. In this paper, we propose a deep learning method for multi-omics integration with incomplete data by Cross-omics Linked unified embedding with Contrastive Learning and Self Attention (CLCLSA). Utilizing complete multi-omics data as supervision, the model employs cross-omics autoencoders to learn the feature representation across different types of biological data. The multi-omics contrastive learning, which is used to maximize the mutual information between different types of omics, is employed before latent feature concatenation. In addition, the feature-level self-attention and omics-level self-attention are employed to dynamically identify the most informative features for multi-omics data integration. Extensive experiments were conducted on four public multi-omics datasets. The experimental results indicated that the proposed CLCLSA outperformed the state-of-the-art approaches for multi-omics data classification using incomplete multi-omics data.

cs.LG↗
arXiv2023

Aggregated (Bi-)Simulation of Finite Valued Networks

The paper provides a method to approximate a large-scale finite-valued network by a smaller model called the aggregated simulation, which is a combination of aggregation and (bi-)simulation. First, the algebraic state space representation (ASSR) of a transition system is presented. Under output equivalence, the quotient system is obtained, which is called the simulation of the original transition system. The ASSR of the quotient system is obtained. The aggregated (bi-)simulation is execueted in several steps: a large scale finite-valued network is firstly aggregated into several blocks, each of which is considered as a network where the in-degree nodes and out-degree nodes are considered as the block inputs and block outputs respectively. Then the dynamics of each block is converted into its quotient system, called its simulation. Then the overall network can be approximated by the quotient systems of each blocks, which is called the aggregated simulation. If the simulation of a block is a bi-simulation, the approximation becomes a lossless transformation. Otherwise, the quotient system is only a (non-deterministic) transition system, and it can be replaced by a probabilistic networks. Aggregated simulation can reduce the dimension of the original network, while a tradeoff between computation complexity and approximation error need to be decided.

eess.SY↗