The clustering-based unsupervised relation discovery method has gradually become one of the important methods of open relation extraction (OpenRE). However, high-dimensional vectors can encode complex linguistic information which leads to the problem
that the derived clusters cannot explicitly align with the relational semantic classes. In this work, we propose a relation-oriented clustering model and use it to identify the novel relations in the unlabeled data. Specifically, to enable the model to learn to cluster relational data, our method leverages the readily available labeled data of pre-defined relations to learn a relation-oriented representation. We minimize distance between the instance with same relation by gathering the instances towards their corresponding relation centroids to form a cluster structure, so that the learned representation is cluster-friendly. To reduce the clustering bias on predefined classes, we optimize the model by minimizing a joint objective on both labeled and unlabeled data. Experimental results show that our method reduces the error rate by 29.2% and 15.7%, on two datasets respectively, compared with current SOTA methods.
We evaluate the light baryonium sepectrum, viz. the baryon-antibaryon states, in the framework of QCD sum rules. The non-perturbative contributions up to dimension 12 are taken into account. Numerical results indicate that there might exist 8 possibl
e light baryonium states, i.e. $p$-$bar{p}$, $Lambda$-$bar{Lambda}$, $Sigma$-$bar{Sigma}$, and $Xi$-$bar{Xi}$ with quantum numbers of $0^{-+}$ and $1^{--}$. For the $Lambda$-$bar{Lambda}$, $Sigma$-$bar{Sigma}$, and $Xi$-$bar{Xi}$ states, their masses are found above the corresponding dibaryon thresholds, while the masses of $p$-$bar{p}$ states are not. The possible baryonium decay modes are analyzed, which are hopefully measurable in BESIII, BELLEII, and LHCb experiments.
We aim at the development and analysis of the numerical schemes for approximately solving the backward diffusion-wave problem, which involves a fractional derivative in time with order $alphain(1,2)$. From terminal observations at two time levels, i.
e., $u(T_1)$ and $u(T_2)$, we simultaneously recover two initial data $u(0)$ and $u_t(0)$ and hence the solution $u(t)$ for all $t > 0$. First of all, existence, uniqueness and Lipschitz stability of the backward diffusion-wave problem were established under some conditions about $T_1$ and $T_2$. Moreover, for noisy data, we propose a quasi-boundary value scheme to regularize the mildly ill-posed problem, and show the convergence of the regularized solution. Next, to numerically solve the regularized problem, a fully discrete scheme is proposed by applying finite element method in space and convolution quadrature in time. We establish error bounds of the discrete solution in both cases of smooth and nonsmooth data. The error estimate is very useful in practice since it indicates the way to choose discretization parameters and regularization parameter, according to the noise level. The theoretical results are supported by numerical experiments.
Toxic comment classification models are often found biased toward identity terms which are terms characterizing a specific group of people such as Muslim and black. Such bias is commonly reflected in false-positive predictions, i.e. non-toxic comment
s with identity terms. In this work, we propose a novel approach to tackle such bias in toxic comment classification, leveraging the notion of subjectivity level of a comment and the presence of identity terms. We hypothesize that when a comment is made about a group of people that is characterized by an identity term, the likelihood of that comment being toxic is associated with the subjectivity level of the comment, i.e. the extent to which the comment conveys personal feelings and opinions. Building upon the BERT model, we propose a new structure that is able to leverage these features, and thoroughly evaluate our model on 4 datasets of varying sizes and representing different social media platforms. The results show that our model can consistently outperform BERT and a SOTA model devised to address identity term bias in a different way, with a maximum improvement in F1 of 2.43% and 1.91% respectively.
The Linked Open Data practice has led to a significant growth of structured data on the Web in the last decade. Such structured data describe real-world entities in a machine-readable way, and have created an unprecedented opportunity for research in
the field of Natural Language Processing. However, there is a lack of studies on how such data can be used, for what kind of tasks, and to what extent they can be useful for these tasks. This work focuses on the e-commerce domain to explore methods of utilising such structured data to create language resources that may be used for product classification and linking. We process billions of structured data points in the form of RDF n-quads, to create multi-million words of product-related corpora that are later used in three different ways for creating of language resources: training word embedding models, continued pre-training of BERT-like language models, and training Machine Translation models that are used as a proxy to generate product-related keywords. Our evaluation on an extensive set of benchmarks shows word embeddings to be the most reliable and consistent method to improve the accuracy on both tasks (with up to 6.9 percentage points in macro-average F1 on some datasets). The other two methods however, are not as useful. Our analysis shows that this could be due to a number of reasons, including the biased domain representation in the structured data and lack of vocabulary coverage. We share our datasets and discuss how our lessons learned could be taken forward to inform future research in this direction.
Emergent cooperative motions of individual degrees of freedom, i.e. collective excitations, govern the low-energy response of system ground states under external stimulations and play essential roles for understanding many-body phenomena in low-dimen
sional materials. The hybridization of distinct collective modes provides a route towards coherent manipulation of coupled degrees of freedom and quantum phases. In magnets, strong coupling between collective spin and lattice excitations, i.e., magnons and phonons, can lead to coherent quasi-particle magnon polarons. Here, we report the direct observation of a series of terahertz magnon polarons in a layered zigzag antiferromagnet FePS3 via far-infrared (FIR) transmission measurements. The characteristic avoided-crossing behavior is clearly seen as the magnon-phonon detuning is continuously changed via Zeeman shift of the magnon mode. The coupling strength g is giant, achieving 120 GHz (0.5 meV), the largest value reported so far. Such a strong coupling leads to a large ratio of g to the resonance frequency (g/{omega}) of 4.5%, and a value of 29 in cooperativity (g^2/{gamma}_{ph}{gamma}_{mag}). Experimental results are well reproduced by first-principle calculations, where the strong coupling is identified to arise from phonon-modulated anisotropic magnetic interactions due to spin-orbit coupling. These findings establish FePS3 as an ideal testbed for exploring hybridization-induced topological magnonics in two dimensions and the coherent control of spin and lattice degrees of freedom in the terahertz regime.
We consider the potential density of rational points on an algebraic variety defined over a number field $K$, i.e., the property that the set of rational points of $X$ becomes Zariski dense after a finite field extension of $K$. For a non-uniruled pr
ojective variety with an int-amplified endomorphism, we show that it always satisfies potential density. When a rationally connected variety admits an int-amplified endomorphism, we prove that there exists some rational curve with a Zariski dense forward orbit, assuming the Zariski dense orbit conjecture in lower dimensions. As an application, we prove the potential density for projective varieties with int-amplified endomorphisms in dimension $leq 3$. We also study the existence of densely many rational points with the maximal arithmetic degree over a sufficiently large number field.
Smart contracts are programs running on blockchain to execute transactions. When input constraints or security properties are violated at runtime, the transaction being executed by a smart contract needs to be reverted to avoid undesirable consequenc
es. On Ethereum, the most popular blockchain that supports smart contracts, developers can choose among three transaction-reverting statements (i.e., require, if...revert, and if...throw) to handle anomalous transactions. While these transaction-reverting statements are vital for preventing smart contracts from exhibiting abnormal behaviors or suffering malicious attacks, there is limited understanding of how they are used in practice. In this work, we perform the first empirical study to characterize transaction-reverting statements in Ethereum smart contracts. We measured the prevalence of these statements in 3,866 verified smart contracts from popular dapps and built a taxonomy of their purposes via manually analyzing 557 transaction-reverting statements. We also compared template contracts and their corresponding custom contracts to understand how developers customize the use of transaction-reverting statements. Finally, we analyzed the security impact of transaction-reverting statements by removing them from smart contracts and comparing the mutated contracts against the original ones. Our study led to important findings, which can shed light on further research in the broad area of smart contract quality assurance and provide practical guidance to smart contract developers on the appropriate use of transaction-reverting statements.
Recently proposed neural architecture search (NAS) methods co-train billions of architectures in a supernet and estimate their potential accuracy using the network weights detached from the supernet. However, the ranking correlation between the archi
tectures predicted accuracy and their actual capability is incorrect, which causes the existing NAS methods dilemma. We attribute this ranking correlation problem to the supernet training consistency shift, including feature shift and parameter shift. Feature shift is identified as dynamic input distributions of a hidden layer due to random path sampling. The input distribution dynamic affects the loss descent and finally affects architecture ranking. Parameter shift is identified as contradictory parameter updates for a shared layer lay in different paths in different training steps. The rapidly-changing parameter could not preserve architecture ranking. We address these two shifts simultaneously using a nontrivial supernet-Pi model, called Pi-NAS. Specifically, we employ a supernet-Pi model that contains cross-path learning to reduce the feature consistency shift between different paths. Meanwhile, we adopt a novel nontrivial mean teacher containing negative samples to overcome parameter shift and model collision. Furthermore, our Pi-NAS runs in an unsupervised manner, which can search for more transferable architectures. Extensive experiments on ImageNet and a wide range of downstream tasks (e.g., COCO 2017, ADE20K, and Cityscapes) demonstrate the effectiveness and universality of our Pi-NAS compared to supervised NAS. See Codes: https://github.com/Ernie1/Pi-NAS.
Multi-view network embedding aims at projecting nodes in the network to low-dimensional vectors, while preserving their multiple relations and attribute information. Contrastive learning-based methods have preliminarily shown promising performance in
this task. However, most contrastive learning-based methods mostly rely on high-quality graph embedding and explore less on the relationships between different graph views. To deal with these deficiencies, we design a novel node-to-node Contrastive learning framework for Multi-view network Embedding (CREME), which mainly contains two contrastive objectives: Multi-view fusion InfoMax and Inter-view InfoMin. The former objective distills information from embeddings generated from different graph views, while the latter distinguishes different graph views better to capture the complementary information between them. Specifically, we first apply a view encoder to generate each graph view representation and utilize a multi-view aggregator to fuse these representations. Then, we unify the two contrastive objectives into one learning objective for training. Extensive experiments on three real-world datasets show that CREME outperforms existing methods consistently.
