Filters








3,575 Hits in 2.4 sec

Multi-Channel Graph Neural Network for Entity Alignment [article]

Yixin Cao, Zhiyuan Liu, Chengjiang Li, Zhiyuan Liu, Juanzi Li, Tat-Seng Chua
2019 arXiv   pre-print
Since it is a Chinese city, KG2 is more informative than KG1 (denoted by dashed lines and ellipse), such as the relations of Dialect and Nearby, and the entity Liu Fei through relation Mayor.  ...  As shown in Figure 1 , to reconcile the differences of Jilin City, it is necessary to complete the missing relations Dialect and Nearby in KG1, and filter out entity Liu Fei exclusive in KG2.  ... 
arXiv:1908.09898v1 fatcat:m3m33i4ukjbv5hye4m6kzzt5uu

Research On The Problems Of Housing Prices In Qingdao From A Macro Perspective

Liu Zhiyuan, Sun Zongdi, Liu Zhiyuan, Sun Zongdi
2016 Zenodo  
Qingdao is a seaside city. Taking into account the characteristics of Qingdao, this article established a multiple linear regression model to analyze the impact of macroeconomic factors on housing prices. We used stepwise regression method to make multiple linear regression analysis, and made statistical analysis of F test values and T test values. According to the analysis results, the model is continuously optimized. Finally, this article obtained the multiple linear regression equation and
more » ... e influencing factors, and the reliability of the model was verified by F test and T test.
doi:10.5281/zenodo.1127507 fatcat:4qnxmjyz3zbvtkcebhrytq7afm

Exploring and Evaluating Attributes, Values, and Structures for Entity Alignment [article]

Zhiyuan Liu, Yixin Cao, Liangming Pan, Juanzi Li, Zhiyuan Liu, Tat-Seng Chua
2021 arXiv   pre-print
Entity alignment (EA) aims at building a unified Knowledge Graph (KG) of rich content by linking the equivalent entities from various KGs. GNN-based EA methods present promising performances by modeling the KG structure defined by relation triples. However, attribute triples can also provide crucial alignment signal but have not been well explored yet. In this paper, we propose to utilize an attributed value encoder and partition the KG into subgraphs to model the various types of attribute
more » ... les efficiently. Besides, the performances of current EA methods are overestimated because of the name-bias of existing EA datasets. To make an objective evaluation, we propose a hard experimental setting where we select equivalent entity pairs with very different names as the test set. Under both the regular and hard settings, our method achieves significant improvements (5.10% on average Hits@1 in DBP15k) over 12 baselines in cross-lingual and monolingual datasets. Ablation studies on different subgraphs and a case study about attribute types further demonstrate the effectiveness of our method. Source code and data can be found at https://github.com/thunlp/explore-and-evaluate.
arXiv:2010.03249v2 fatcat:mq4xitmoevc5rnszdimxkvqxmm

Comprehend DeepWalk as Matrix Factorization [article]

Cheng Yang, Zhiyuan Liu
2015 arXiv   pre-print
Word2vec, as an efficient tool for learning vector representation of words has shown its effectiveness in many natural language processing tasks. Mikolov et al. issued Skip-Gram and Negative Sampling model for developing this toolbox. Perozzi et al. introduced the Skip-Gram model into the study of social network for the first time, and designed an algorithm named DeepWalk for learning node embedding on a graph. We prove that the DeepWalk algorithm is actually factoring a matrix M where each
more » ... y M_ij is logarithm of the average probability that node i randomly walks to node j in fix steps.
arXiv:1501.00358v1 fatcat:dhcwmgyewbblviy2wghekq7r24

Explore Entity Embedding Effectiveness in Entity Retrieval [article]

Zhenghao Liu, Chenyan Xiong, Maosong Sun, Zhiyuan Liu
2019 arXiv   pre-print
This paper explores entity embedding effectiveness in ad-hoc entity retrieval, which introduces distributed representation of entities into entity retrieval. The knowledge graph contains lots of knowledge and models entity semantic relations with the well-formed structural representation. Entity embedding learns lots of semantic information from the knowledge graph and represents entities with a low-dimensional representation, which provides an opportunity to establish interactions between
more » ... related entities and candidate entities for entity retrieval. Our experiments demonstrate the effectiveness of entity embedding based model, which achieves more than 5\% improvement than the previous state-of-the-art learning to rank based entity retrieval model. Our further analysis reveals that the entity semantic match feature effective, especially for the scenario which needs more semantic understanding.
arXiv:1908.10554v1 fatcat:zepi5ixl3bfgfnqz5zriodjhtq

More Robust Dense Retrieval with Contrastive Dual Learning [article]

Yizhi Li, Zhenghao Liu, Chenyan Xiong, Zhiyuan Liu
2021 arXiv   pre-print
Dense retrieval conducts text retrieval in the embedding space and has shown many advantages compared to sparse retrieval. Existing dense retrievers optimize representations of queries and documents with contrastive training and map them to the embedding space. The embedding space is optimized by aligning the matched query-document pairs and pushing the negative documents away from the query. However, in such training paradigm, the queries are only optimized to align to the documents and are
more » ... rsely positioned, leading to an anisotropic query embedding space. In this paper, we analyze the embedding space distributions and propose an effective training paradigm, Contrastive Dual Learning for Approximate Nearest Neighbor (DANCE) to learn fine-grained query representations for dense retrieval. DANCE incorporates an additional dual training object of query retrieval, inspired by the classic information retrieval training axiom, query likelihood. With contrastive learning, the dual training object of DANCE learns more tailored representations for queries and documents to keep the embedding space smooth and uniform, thriving on the ranking performance of DANCE on the MS MARCO document retrieval task. Different from ANCE that only optimized with the document retrieval task, DANCE concentrates the query embeddings closer to document representations while making the document distribution more discriminative. Such concentrated query embedding distribution assigns more uniform negative sampling probabilities to queries and helps to sufficiently optimize query representations in the query retrieval task. Our codes are released at https://github.com/thunlp/DANCE.
arXiv:2107.07773v1 fatcat:7i4ngfxk5zcjnkit2tbnmdzcdi

Lifelong Learning for Sentiment Classification [article]

Zhiyuan Chen, Nianzu Ma, Bing Liu
2018 arXiv   pre-print
Liu (2012) and Pang and Lee (2008) provided good surveys of the existing research.  ...  Existing lifelong learning approaches focused on exploiting invariances (Thrun, 1998) and other types of knowledge (Chen and Liu, 2014b , Chen and Liu, 2014a , Ruvolo and Eaton, 2013 across multiple  ... 
arXiv:1801.02808v1 fatcat:eek4xq7ybfb5dgxlcsc6qu6mw4

Multi-Channel Graph Neural Network for Entity Alignment

Yixin Cao, Zhiyuan Liu, Chengjiang Li, Zhiyuan Liu, Juanzi Li, Tat-Seng Chua
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
Since it is a Chinese city, KG2 is more informative than KG1 (denoted by dashed lines and ellipse), such as the relations of Dialect and Nearby, and the entity Liu Fei through relation Mayor.  ...  As shown in Figure 1 , to reconcile the differences of Jilin City, it is necessary to complete the missing relations Dialect and Nearby in KG1, and filter out entity Liu Fei exclusive in KG2.  ... 
doi:10.18653/v1/p19-1140 dblp:conf/acl/CaoLLLLC19 fatcat:7gzm33p5bbdf3aum5gh4z5oe3q

Few-Shot Generative Conversational Query Rewriting [article]

Shi Yu, Jiahua Liu, Jingqin Yang, Chenyan Xiong, Paul Bennett, Jianfeng Gao, Zhiyuan Liu
2020 arXiv   pre-print
Conversational query rewriting aims to reformulate a concise conversational query to a fully specified, context-independent query that can be effectively handled by existing information retrieval systems. This paper presents a few-shot generative approach to conversational query rewriting. We develop two methods, based on rules and self-supervised learning, to generate weak supervision data using large amounts of ad hoc search sessions, and to fine-tune GPT-2 to rewrite conversational queries.
more » ... n the TREC Conversational Assistance Track, our weakly supervised GPT-2 rewriter improves the state-of-the-art ranking accuracy by 12%, only using very limited amounts of manual query rewrites. In the zero-shot learning setting, the rewriter still gives a comparable result to previous state-of-the-art systems. Our analyses reveal that GPT-2 effectively picks up the task syntax and learns to capture context dependencies, even for hard cases that involve group references and long-turn dependencies.
arXiv:2006.05009v1 fatcat:bxqbdlqh5vgbbi3j2nevbodv5m

Neural Diffusion Model for Microscopic Cascade Prediction [article]

Cheng Yang, Maosong Sun, Haoran Liu, Shiyi Han, Zhiyuan Liu, Huanbo Luan
2018 arXiv   pre-print
The prediction of information diffusion or cascade has attracted much attention over the last decade. Most cascade prediction works target on predicting cascade-level macroscopic properties such as the final size of a cascade. Existing microscopic cascade prediction models which focus on user-level modeling either make strong assumptions on how a user gets infected by a cascade or limit themselves to a specific scenario where "who infected whom" information is explicitly labeled. The strong
more » ... mptions oversimplify the complex diffusion mechanism and prevent these models from better fitting real-world cascade data. Also, the methods which focus on specific scenarios cannot be generalized to a general setting where the diffusion graph is unobserved. To overcome the drawbacks of previous works, we propose a Neural Diffusion Model (NDM) for general microscopic cascade prediction. NDM makes relaxed assumptions and employs deep learning techniques including attention mechanism and convolutional network for cascade modeling. Both advantages enable our model to go beyond the limitations of previous methods, better fit the diffusion data and generalize to unseen cascades. Experimental results on diffusion prediction task over four realistic cascade datasets show that our model can achieve a relative improvement up to 26% against the best performing baseline in terms of F1 score.
arXiv:1812.08933v1 fatcat:cmix6xmcg5aqbgyxb3x6kvo2qe

Linking GloVe with word2vec [article]

Tianze Shi, Zhiyuan Liu
2014 arXiv   pre-print
The Global Vectors for word representation (GloVe), introduced by Jeffrey Pennington et al. is reported to be an efficient and effective method for learning vector representations of words. State-of-the-art performance is also provided by skip-gram with negative-sampling (SGNS) implemented in the word2vec tool. In this note, we explain the similarities between the training objectives of the two models, and show that the objective of SGNS is similar to the objective of a specialized form of
more » ... , though their cost functions are defined differently.
arXiv:1411.5595v2 fatcat:ugi5gnfxdjbg7egwnvd3vnawki

Additive manufacturing of thin electrolyte layers via inkjet printing of highly-stable ceramic inks

Zhongqi Zhu, Zhiyuan Gong, Piao Qu, Ziyong Li, Sefiu Abolaji Rasaki, Zhiyuan Liu, Pei Wang, Changyong Liu, Changshi Lao, Zhangwei Chen
2021 Journal of Advanced Ceramics  
AbstractInkjet printing is a promising alternative for the fabrication of thin film components for solid oxide fuel cells (SOFCs) due to its contactless, mask free, and controllable printing process. In order to obtain satisfying electrolyte thin layer structures in anode-supported SOFCs, the preparation of suitable electrolyte ceramic inks is a key. At present, such a kind of 8 mol% Y2O3-stabilized ZrO2 (8YSZ) electrolyte ceramic ink with long-term stability and high solid loading (> 15 wt%)
more » ... ems rare for precise inkjet printing, and a number of characterization and performance aspects of the inks, such as homogeneity, viscosity, and printability, should be studied. In this study, 8YSZ ceramic inks of varied compositions were developed for inkjet printing of SOFC ceramic electrolyte layers. The dispersing effect of two types of dispersants, i.e., polyacrylic acid ammonium (PAANH4) and polyacrylic acid (PAA), were compared. The results show that ultrasonic dispersion treatment can help effectively disperse the ceramic particles in the inks. PAANH4 has a better dispersion effect for the inks developed in this study. The inks show excellent printable performance in the actual printing process. The stability of the ink can be maintained for a storage period of over 30 days with the help of initial ultrasonic dispersion. Finally, micron-size thin 8YSZ electrolyte films were successfully fabricated through inkjet printing and sintering, based on the as-developed high solid loading 8YSZ inks (20 wt%). The films show fully dense and intact structural morphology and smooth interfacial bonding, offering an improved structural quality of electrolyte for enhanced SOFC performance.
doi:10.1007/s40145-020-0439-9 fatcat:wij2ydvlpzbtbjeh7gztbty54u

Multi-Paragraph Reasoning with Knowledge-enhanced Graph Neural Network [article]

Deming Ye, Yankai Lin, Zhenghao Liu, Zhiyuan Liu, Maosong Sun
2019 arXiv   pre-print
Multi-paragraph reasoning is indispensable for open-domain question answering (OpenQA), which receives less attention in the current OpenQA systems. In this work, we propose a knowledge-enhanced graph neural network (KGNN), which performs reasoning over multiple paragraphs with entities. To explicitly capture the entities' relatedness, KGNN utilizes relational facts in knowledge graph to build the entity graph. The experimental results show that KGNN outperforms in both distractor and full wiki
more » ... settings than baselines methods on HotpotQA dataset. And our further analysis illustrates KGNN is effective and robust with more retrieved paragraphs.
arXiv:1911.02170v1 fatcat:2fzzx7vjxjhbviy2twmhj7l7xa

TIAGE: A Benchmark for Topic-Shift Aware Dialog Modeling [article]

Huiyuan Xie, Zhenghao Liu, Chenyan Xiong, Zhiyuan Liu, Ann Copestake
2021 arXiv   pre-print
Zhenghao Liu is supported by National Natural Science Foundation of China (NSFC) under grant No. 61872074 and 61772122.  ...  ., 2019; Liu et al., 2018) have been reported to perform well in generating on-topic utterances in dialog scenarios.  ...  Task-oriented dialog systems (Budzianowski et al., 2018; Liu et al., 2018) help users complete tasks in specific domains.  ... 
arXiv:2109.04562v1 fatcat:ejrrv5v54zctrgjgi72yemmbhq

Topic Sensitive Neural Headline Generation [article]

Lei Xu, Ziyun Wang, Ayana, Zhiyuan Liu, Maosong Sun
2016 arXiv   pre-print
Neural models have recently been used in text summarization including headline generation. The model can be trained using a set of document-headline pairs. However, the model does not explicitly consider topical similarities and differences of documents. We suggest to categorizing documents into various topics so that documents within the same topic are similar in content and share similar summarization patterns. Taking advantage of topic information of documents, we propose topic sensitive
more » ... al headline generation model. Our model can generate more accurate summaries guided by document topics. We test our model on LCSTS dataset, and experiments show that our method outperforms other baselines on each topic and achieves the state-of-art performance.
arXiv:1608.05777v1 fatcat:yqqicm6skfdwnco54ohssuxzpq
« Previous Showing results 1 — 15 out of 3,575 results