Illinois-LH: A Denotational and Distributional Approach to Semantics

By Alice Lai, Julia Hockenmaier. 2014

132 citations recorded by the source corpus.

This paper describes and analyzes our SemEval 2014 Task 1 system. Its features are based on distributional and denotational similarities; word alignment; negation; and hypernym/hyponym, synonym, and antonym relations.

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

Supervised Learning of Universal Sentence Representations from Natural Language Inference Data (2017, neutral): “9% test accuracy on SICK-E while previous best handengineered models (Lai and Hockenmaier, 2014) obtained 84.”
A large annotated corpus for learning natural language inference (2015, background): “6% (Lai and Hockenmaier, 2014) and the 84% level of interannotator agreement, which likely represents an approximate performance ceiling.”
Skip-Thought Vectors (2015, neutral): “Illinois-LH [18] 0.”
Annotation Artifacts in Natural Language Inference Data (2018, background): “Levy and Dagan (2016) showed that a relation inference benchmark (Zeichner et al.”
SentEval: An Evaluation Toolkit for Universal Sentence Representations (2018, background): “ined directly on each task (no transfer). Results 1 correspond to AdaSent (Zhao et al., 2015), 2 to BLSTM-2DCNN (Zhou et al., 2016), 3 to TF-KLD (Ji and Eisenstein, 2013) and 4 to Illinois-LH system (Lai and Hockenmaier, 2014). • dropout(ﬂoat): dropout rate in the case of MLP. For use cases where th”
Probing for semantic evidence of composition by means of simple classification tasks (2016, neutral): “…et al. (2016) point out, 86.4% of sentence pairs labeled as CON-TRADICTION can be identiﬁed simply by detecting the presence of negation; a similar obser-134 vation is made by Lai and Hockenmaier (2014)), which means that we cannot draw ﬁrm composition conclusions from performance on this task.”
A Joint Many-Task Model: Growing a Neural Network for Multiple NLP Tasks (2016, neutral): “2 Lai and Hockenmaier (2014) 84.”
Reasoning about Entailment with Neural Attention (2015, background): “…systems for RTE so far relied heavily on engineered NLP pipelines, extensive manual creation of features, as well as various external resources and specialized subcomponents such as negation detection (e.g. Lai and Hockenmaier, 2014; Jimenez et al., 2014; Zhao et al., 2014; Beltagy et al., 2015).”
SemEval-2014 Task 1: Evaluation of Compositional Distributional Semantic Models on Full Sentences through Semantic Relatedness and Textual Entailment (2014, background): “The fine-grained analysis reported by IllinoisLH (Lai and Hockenmaier, 2014) shows that a full compositional system (based on pointwise multiplication) fails to capture contradiction.”
Multi-Reward Reinforced Summarization with Saliency and Entailment (2018, methodology): “Recognizing Textual Entailment (RTE), the task of classifying two sentences as entailment, contra-diction, or neutral, has been used for Q&A and IE tasks (Harabagiu and Hickl, 2006; Dagan et al., 2006; Lai and Hockenmaier, 2014; Jimenez et al., 2014).”
ABCNN: Attention-Based Convolutional Neural Network for Modeling Sentence Pairs (2015, methodology): “Following Lai and Hockenmaier (2014), we use WordNet to detect synonyms, hypernyms and antonyms in the pairs.”
Reinforced Video Captioning with Entailment Rewards (2017, background): “Recognizing Textual Entailment (RTE) is a traditional NLP task (Dagan et al., 2006; Lai and Hockenmaier, 2014; Jimenez et al., 2014), boosted by a large dataset (SNLI) recently introduced by Bowman et al. (2015).”
Jointly optimizing word representations for lexical and sentential tasks with the C-PHRASE model (2015, background): “The current state-of-the-art is reached by Lai and Hockenmaier (2014), using a much richer”
Siamese Recurrent Architectures for Learning Sentence Similarity (2016, neutral): “Table 4 shows that such an approach outperforms all other textual-entailment systems except for the Illinois-LH system of Lai and Hockenmaier (2014).”
ParaNMT-50M: Pushing the Limits of Paraphrastic Sentence Embeddings with Millions of Machine Translations (2017, neutral): “rvised (Direct) Naive Bayes - SVM 79.4 81.8 93.2 86.3 83.1 - - - - AdaSent (Zhao et al., 2015) 83.1 86.3 95.5 93.3 - 92.4 - - - TF-KLD (Ji and Eisenstein, 2013) - - - - - - 80.4/85.9 - - Illinois-LH (Lai and Hockenmaier, 2014) - - - - - - - - 84.5 Dependency Tree-LSTM (Tai et al., 2015) - - - - - - ”
Pairwise Word Interaction Modeling with Deep Neural Networks for Semantic Similarity Measurement (2016, neutral): “We compared our model to competitive baselines prepared by Yang et al. (2015) and also evaluated He et al.”