Natural Language Inference over Interaction Space

By Yichen Gong, Heng Luo, Jian Zhang. 2017

279 citations recorded by the source corpus.

Natural Language Inference (NLI) task requires an agent to determine the logical relationship between a natural language premise and a natural language hypothesis. We introduce Interactive Inference Network (IIN), a novel class of neural network architectures that is able to achieve high-level understanding of the sentence pair by hierarchically extracting semantic features from interaction space. We show that an interaction tensor (attention weight) contains semantic information to solve natural language inference, and a denser interaction tensor contains richer semantic information. One instance of such architecture, Densely Interactive Inference Network (DIIN), demonstrates the state-of-the-art performance on large scale NLI copora and large-scale NLI alike corpus. It's noteworthy that DIIN achieve a greater than 20% error reduction on the challenging Multi-Genre NLI (MultiNLI) dataset with respect to the strongest published system.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

Dissecting Contextual Word Embeddings: Architecture and Representation (2018, neutral): “9% mismatched from Gong et al. (2018).”
GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding (2018, methodology): “MNLI The Multi-Genre Natural Language Inference Corpus (Williams et al., 2018) is a crowd-sourced collection of sentence pairs with textual entailment annotations.”
Deep Contextualized Word Representations (2018, background): “The augmented con-Model Feature based (Bowman et al., 2015) 78.2 DIIN (Gong et al., 2018) 88.0 BCN+Char+CoVe (McCann et al., 2017) 88.1 ESIM (Chen et al., 2017) 88.0 ESIM+TreeLSTM (Chen et al., 2017) 88.6 ESIM+ELMo 88.7 ± 0.17 DIIN ensemble (Gong et al., 2018) 88.9 ESIM+ELMo ensemble 89.3 text vecto”
Adversarial NLI: A New Benchmark for Natural Language Understanding (2019, background): “…from the original papers: for SNLI-Hard from Gururangan et al.’s implementation of the hierarchical tensor-based Densely Inter-active Inference Network (Gong et al., 2018, DIIN) on MNLI, and for the NLI stress tests, Naik et al.’s implementation of InferSent (Conneau et al., 2017) trained on SNLI.”
XNLI: Evaluating Cross-lingual Sentence Representations (2018, background): “…a million examples, and these have been widely used to evaluate neural network architectures and training strategies (Rocktäschel et al., 2016; Gong et al., 2018; Peters et al., 2018; Wang et al., 2018), as well as to train effective, reusable sentence representations (Conneau et al., 2017;…”
e-SNLI: Natural Language Inference with Natural Language Explanations (2018, background): “Attention mechanisms in neural networks brought consistent improvements over the non-attention counter-parts in various areas, such as computer vision [27], speech [4], or natural language processing [12, 2].”
BERTScore: Evaluating Text Generation with BERT (2019, neutral): “We also compare our correlations with state-of-the-art metrics, including METEOR++ (Guo et al., 2018), CHR F++ (Popovi´c, 2017), BEER (Stano-jevi´c and Sima’an, 2014), BLEND (Ma et al., 2017), and RUSE (Shimanaka et al., 2018).”
PAWS: Paraphrase Adversaries from Word Scrambling (2019, methodology): “On PAWS Wiki , this addition lifts the (no pre-training) DIIN model AUC from 91.1% (Table 8) to 93.8% (Table 9).”
Combining Fact Extraction and Verification with Neural Semantic Matching Networks (2018, background): “…Language Inference (Williams, Nangia, and Bowman 2018) (Multi-NLI) have promoted the development of many different neural NLI models (Nie and Bansal 2017; Conneau et al. 2017; Parikh et al. 2016; Chen et al. 2017b; Gong, Luo, and Zhang 2017; Ghaeini et al. 2018) that achieve promising performance.”
End-to-End Retrieval in Continuous Space (2018, background): “First, a re-ranking model has access to the raw representations of both query and candidate and can thus learn complex interactions (Parikh et al., 2016; Gong et al., 2017), whereas a retrieval model must encode queries and candidates independently to allow for fast neighbor look-up.”
Convolutional Interaction Network for Natural Language Inference (2018, unknown): “…Match-LSTM (Wang and Jiang, 2016) 92.0 86.1 Decomposable attention model (Parikh et al., 2016) 90.5 86.8 BiMPM (Zhiguo Wang, 2017) 90.9 87.5 NTI-SLSTM-LSTM (Munkhdalai and Yu, 2017) 88.5 87.3 Re-read LSTM (Sha et al., 2016) 90.7 87.5 DIIN (Gong et al., 2017) 91.2 88.0 ESIM (Chen et al., 2017a) 92”
Interpreting Recurrent and Attention-Based Neural Models: a Case Study on Natural Language Inference (2018, methodology): “Many of the top performing NLI models (Ghaeini et al., 2018b; Tay et al., 2018; Peters et al., 2018; McCann et al., 2017; Gong et al., 2017; Wang et al., 2017; Chen et al., 2017), are variants of the ESIM model (Chen et al.”
The BQ Corpus: A Large-scale Domain-specific Chinese Corpus For Sentence Semantic Equivalence Identification (2018, background): “The Accuracy of BiMPM is 81.85% and that of DIIN is 81.41%.”
Raise a Child in Large Language Model: Towards Effective and Generalizable Fine-tuning (2021, methodology): “We follow Gong et al. (2018) and use the same mapping strategy as SciTail.”
Learning Visually-Grounded Semantics from Contrastive Adversarial Samples (2018, methodology): “Training Inspired by Gong et al. (2018), we train an image-word alignment network through the interaction space, since this structure reﬂects the property of “alignment” better than just concatenating the feature vectors of word and image.”
AdvEntuRe: Adversarial Training for Textual Entailment with Knowledge-Guided Examples (2018, background): “On certain datasets, recent deep learning entailment systems (Parikh et al., 2016; Wang et al., 2017; Gong et al., 2018) have achieved close to human level performance.”