Sentence embeddings in NLI with iterative refinement encoders

Talman, Aarne; Yli-Jyrä, Anssi; Tiedemann, Jörg

doi:10.1017/s1351324919000202

Cited by 37 publications

(23 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where T NLI can be any of the existing state of the art textbased NLI models (Wang and Jiang 2016a;Talman, Yli-Jyrä, and Tiedemann 2019;Liu et al 2019a).…”

Section: A Standard Text-based Modelmentioning

confidence: 99%

Infusing Knowledge into the Textual Entailment Task Using Graph Convolutional Networks

Kapanipathi¹,

Thost²,

Patel³

et al. 2020

AAAI

View full text Add to dashboard Cite

Textual entailment is a fundamental task in natural language processing. Most approaches for solving this problem use only the textual content present in training data. A few approaches have shown that information from external knowledge sources like knowledge graphs (KGs) can add value, in addition to the textual content, by providing background knowledge that may be critical for a task. However, the proposed models do not fully exploit the information in the usually large and noisy KGs, and it is not clear how it can be effectively encoded to be useful for entailment. We present an approach that complements text-based entailment models with information from KGs by (1) using Personalized PageRank to generate contextual subgraphs with reduced noise and (2) encoding these subgraphs using graph convolutional networks to capture the structural and semantic information in KGs. We evaluate our approach on multiple textual entailment datasets and show that the use of external knowledge helps the model to be robust and improves prediction accuracy. This is particularly evident in the challenging BreakingNLI dataset, where we see an absolute improvement of 5-20% over multiple text-based entailment models.

show abstract

“…where T NLI can be any of the existing state of the art textbased NLI models (Wang and Jiang 2016a;Talman, Yli-Jyrä, and Tiedemann 2019;Liu et al 2019a).…”

Section: A Standard Text-based Modelmentioning

confidence: 99%

Infusing Knowledge into the Textual Entailment Task Using Graph Convolutional Networks

Kapanipathi¹,

Thost²,

Patel³

et al. 2020

AAAI

View full text Add to dashboard Cite

show abstract

“…The Premise-Hypothesis pairs were created utilizing about 3500 multiplechoice queries with no or a little involvement of annotators with annotating the pairs. They investigated several techniques ranging from the traditional ones to the state-ofthe-art methods bundling word embedding methods including word2vec [10], fastText [11], ELMo [12], BERT [13], and LASER [14], various modeling approaches specifically dedicated to Natural Language Inference dealing, such as, DecompAtt [15], ESIM [16], HBMP [17], ULMFiT [18], and cross-lingual transfer approaches. Another research conducted by Mohammad Mosharaf Hossain et al [19] dealt with countering negation within English sentences as they are ubiquitous in common English sentences.…”

Section: Related Workmentioning

confidence: 99%

UT-ATD: Universal Transformer for Anomalous Trajectory Detection by Embedding Trajectory Information

Zhang¹

2021

JVLC

View full text Add to dashboard Cite

Natural Language Interface or NLI has the potential to add syllogistic reasoning over the already existing facts and develop a new kind of knowledge dataset in itself. In this paper, we have demonstrated a Recognizing Textual Entailment wherein the task is to recognize whether a given hypothesis is true (Entailment), false (Contradiction) or unrelated(neutral) with respect to the sentence called premise. The task is performed by training MNLI corpus along with the manually collected dataset from Amazon Product Reviews each having hypothesis and premise pairs with corresponding labels. With this use case, we propose to bring sustainable development in the classification methods used by major E-commerce companies.

show abstract

“…There are many methods used in text or sentence representations. These methods generally first translate words into word embeddings through a projection layer and then combine such embeddings with different architectures such as Recurrent Neural Network (RNN) [32]- [34], Convolutional Neural Network (CNN) [35], [36]. In this paper, to better encode the semantics of entity description, we focus on the sentence embedding approach.…”

Section: B Hierarchical Bilstm Max Pooling Encodermentioning

confidence: 99%

Knowledge Graph Embedding With Interactive Guidance From Entity Descriptions

2019

View full text Add to dashboard Cite

Knowledge Graph (KG) embedding aims to represent both entities and relations into a continuous low-dimensional vector space. Most previous attempts perform the embedding task using only knowledge triples to indicate relations between entities. Entity descriptions, although containing rich background information, have not been well utilized in these methods. In this paper, we propose Entity Descriptions-Guided Embedding (EDGE), a novel method for learning the knowledge graph representations with semantic guidance from entity descriptions. EDGE enables an embedding model to learn simultaneously from 1) knowledge triples that have been directly observed in a given KG, and 2) entity descriptions which have rich semantic information about these entities. In the learning process, EDGE encodes the semantics of entity descriptions to enhance the learning of knowledge graph embedding, and integrates such learned KG embedding to constraint their corresponding word embeddings in entity descriptions. Through this interactive procedure, semantics of entity descriptions may be better transferred into the learned KG embedding. We evaluate EDGE in link prediction and entity classification on Freebase and WordNet. Experimental results show that: 1) with entity descriptions injected, EDGE achieves significant and consistent improvements over state-of-the-art baselines; and 2) compared to those one-time injection schemes studied before, the interactive guidance strategy maximizes the utility of entity descriptions for KG embedding, and indeed achieves substantially better performance.INDEX TERMS Knowledge graph embedding, entity descriptions, interactive guidance.

show abstract

Sentence embeddings in NLI with iterative refinement encoders

Cited by 37 publications

References 21 publications

Infusing Knowledge into the Textual Entailment Task Using Graph Convolutional Networks

Infusing Knowledge into the Textual Entailment Task Using Graph Convolutional Networks

UT-ATD: Universal Transformer for Anomalous Trajectory Detection by Embedding Trajectory Information

Knowledge Graph Embedding With Interactive Guidance From Entity Descriptions

Contact Info

Product

Resources

About