AliCG: Fine-grained and Evolvable Conceptual Graph Construction for Semantic Search at Alibaba

Zhang, Ningyu; Jia, Qianghuai; Chen, Xiang; Ye, Hongbin; Chen, Hui; Tou, Huaixiao; Huang, Gang; Wang, Zhao; Hua, Nengwei; Chen, Huajun

doi:10.48550/arxiv.2106.01686

Cited by 3 publications

(1 citation statement)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Living in an enormous web connecting us to each other, we now face a new reality: it is very hard to escape one's past on the Internet now that every photo, status update, and tweet lives forever in the cloud [19,67,81,85]. Moreover, recommender systems that actively explore the user data [22,100] for data-driven algorithms have encouraged the debate about the right to privacy over convenience. Fortunately, we have the Right to Be Forgotten (RTBF), which gives individuals the right to ask organizations to delete their personal data.…”

Section: Introductionmentioning

confidence: 99%

Burn After Reading: Online Adaptation for Cross-domain Streaming Data

Yang¹,

Gao²,

Chen³

et al. 2021

Preprint

View full text Add to dashboard Cite

In the context of online privacy, many methods propose complex privacy and security preserving measures to protect sensitive data. In this paper we argue that: not storing any sensitive data is the best form of security. Thus we propose an online framework that "burns after reading", i.e. each online sample is immediately deleted after it is processed. Meanwhile, we tackle the inevitable distribution shift between the labeled public data and unlabeled private data as a problem of unsupervised domain adaptation. Specifically, we propose a novel algorithm that aims at the most fundamental challenge of the online adaptation setting-the lack of diverse source-target data pairs. Therefore, we design a Cross-Domain Bootstrapping approach, called CRODOBO, to increase the combined diversity across domains. Further, to fully exploit the valuable discrepancies among the diverse combinations, we employ the training strategy of multiple learners with cosupervision. CRODOBO achieves state-of-the-art online performance on four domain adaptation benchmarks.

show abstract

Section: Introductionmentioning

confidence: 99%

Burn After Reading: Online Adaptation for Cross-domain Streaming Data

Yang¹,

Gao²,

Chen³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

Drop Redundant, Shrink Irrelevant: Selective Knowledge Injection for Language Pretraining

Zhang

Cheng²,

Chen

et al. 2021

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence

View full text Add to dashboard Cite

Previous research has demonstrated the power of leveraging prior knowledge to improve the performance of deep models in natural language processing. However, traditional methods neglect the fact that redundant and irrelevant knowledge exists in external knowledge bases. In this study, we launched an in-depth empirical investigation into downstream tasks and found that knowledge-enhanced approaches do not always exhibit satisfactory improvements. To this end, we investigate the fundamental reasons for ineffective knowledge infusion and present selective injection for language pretraining, which constitutes a model-agnostic method and is readily pluggable into previous approaches. Experimental results on benchmark datasets demonstrate that our approach can enhance state-of-the-art knowledge injection methods.

show abstract

Speaker-Oriented Latent Structures for Dialogue-Based Relation Extraction

Nan¹,

Luo²,

Leng³

et al. 2021

Preprint

View full text Add to dashboard Cite

Dialogue-based relation extraction (DiaRE)aims to detect the structural information from unstructured utterances in dialogues. Existing relation extraction models may be unsatisfactory under such a conversational setting, due to the entangled logic and information sparsity issues in utterances involving multiple speakers. To this end, we introduce SOLS, a novel model which can explicitly induce speaker-oriented latent structures for better DiaRE. Specifically, we learn latent structures to capture the relationships among tokens beyond the utterance boundaries, alleviating the entangled logic issue. During the learning process, our speakerspecific regularization method progressively highlights speaker-related key clues and erases the irrelevant ones, alleviating the information sparsity issue. Experiments on three public datasets demonstrate the effectiveness of our proposed approach. * * Equal contribution, work done during Guoqing Luo and Yao Xiao's internships at SUTD.Accepted as a long paper in the main conference of EMNLP 2021 (Conference on Empirical Methods in Natural Language Processing). S1: Jack. Could you come in here for a moment? Now! S2: Found it. S3: I'll take that dad. S1: It seems your daughter and Richard are something of an item. S2: That's impossible, he's got a twinkie in the city. S4: Dad, I'm the twinkie. S2: You're the twinkie? S3: Yes, that is impossible S5: She's not a twinkie.S2: Am I supposed to stand here and listen to this on my birthday? S4: Dad, dad this is a good thing for me. Ya know, and you even said yourself, you've never seen Richard happier.

show abstract

AliCG: Fine-grained and Evolvable Conceptual Graph Construction for Semantic Search at Alibaba

Cited by 3 publications

References 20 publications

Burn After Reading: Online Adaptation for Cross-domain Streaming Data

Burn After Reading: Online Adaptation for Cross-domain Streaming Data

Drop Redundant, Shrink Irrelevant: Selective Knowledge Injection for Language Pretraining

Speaker-Oriented Latent Structures for Dialogue-Based Relation Extraction

Contact Info

Product

Resources

About