An extended model of natural logic

By Bill MacCartney, Christopher D. Manning. 2009

238 citations recorded by the source corpus.

We propose a model of natural language inference which identifies valid inferences by their lexical and syntactic features, without full semantic interpretation. We extend past work in natural logic, which has focused on semantic containment and monotonicity, by incorporating both semantic exclusion and implicativity. Our model decomposes an inference problem into a sequence of atomic edits linking premise to hypothesis; predicts a lexical semantic relation for each edit; propagates these relations upward through a semantic composition tree according to properties of intermediate nodes; and joins the resulting semantic relations across the edit sequence. A computational implementation of the model achieves 70% accuracy and 89% precision on the FraCaS test suite. Moreover, including this model as a component in an existing system yields significant performance gains on the Recognizing Textual Entailment challenge.

Subfields: Artificial Intelligence, Molecular Biology, 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 Molecular Biology

In Natural Language Processing

Citations in the corpus

A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference (2017, background): “Nine of these come from the second release of the Open American National Corpus (OANC; Fill-more et al., 1998; Macleod et al., 2000; Ide and Suderman, 2006, downloaded 12/2016): • F ACE -”
A large annotated corpus for learning natural language inference (2015, background): “Thus, natural language inference (NLI) — characterizing and using these relations in computational systems (Fyodorov et al., 2000; Condoravdi et al., 2003; Bos and Markert, 2005; Dagan et al., 2006; MacCartney and Manning, 2009) — is essential in tasks ranging from information retrieval to semantic ”
The Natural Language Decathlon: Multitask Learning as Question Answering (2018, background): “Natural Language Inference NLI has a long history playing roles in tasks like information retrieval and semantic parsing [Fyodorov et al., 2000, Condoravdi et al., 2003, Bos and Markert, 2005, Dagan et al., 2005, MacCartney and Manning, 2009].”
RuleBERT: Teaching Soft Rules to Pre-Trained Language Models (2021, background): “Yet, we are also different from Natural Language Inference (NLI) and textual entailment, which work with text directly, but cannot handle Horn rules (MacCartney and Manning, 2009; Dagan et al., 2013).”
Visual Entailment: A Novel Task for Fine-Grained Image Understanding (2019, methodology): “We focus on textual entailment as our NLI task [18, 11, 3, 12, 46].”
Natural Language Inference over Interaction Space (2017, methodology): “The previous work on NLI (or RTE) has extensively researched on conventional approaches(Fyodorov et al., 2000; Bos & Markert, 2005; MacCartney & Manning, 2009).”
Abductive Commonsense Reasoning (2019, background): “Related Datasets Our new resource ART complements ongoing efforts in building resources for natural language inference (Dagan et al., 2006; MacCartney & Manning, 2009; Bowman et al., 2015; Williams et al., 2018a; Camburu et al., 2018).”
Performance Impact Caused by Hidden Bias of Training Data for Recognizing Textual Entailment (2018, methodology): “This paper proposes a new empirical method to investigate a quality of a large corpus designed for the recognizing textual entailment (RTE) task (Condoravdi et al., 2003; Bos and Markert, 2005; MacCartney and Manning, 2009; Marelli et al., 2014a).”
Dialogue Natural Language Inference (2018, methodology): “Separately, the framework of Natural Language Inference (NLI) [2, 5, 14] involves learning a mapping between a sentence pair and an entailment category.”
Tree-Structured Composition in Neural Networks without Tree-Structured Architectures (2015, methodology): “The data we use are labeled with the seven mutually exclusive logical relations of [11], which distinguish entailment in two directions (@, A), equivalence (≡), exhaustive and non-exhaustive contradiction (∧, |), and two types of semantic independence (#, `).”
Analyzing Dynamic Adversarial Training Data in the Limit (2021, neutral): “We choose to focus on NLI, a canonical and wellstudied natural language understanding task (Dagan et al., 2005; Bos and Markert, 2005; Giampiccolo et al., 2007; MacCartney and Manning, 2009).”
Recursive Neural Networks Can Learn Logical Semantics (2014, methodology): “Proceedings of the 3rd Workshop on Continuous Vector Space Models and their Compositionality (CVSC), pages 12–21, Beijing, China, July 26-31, 2015. c©2015 Association for Computational Linguistics”
Siamese recurrent networks learn first-order logic reasoning and exhibit zero-shot compositional generalization (2019, background): “Table 2: The seven entailment relations of MacCartney and Manning (2009).”
Recent Progress on Monotonicity (2014, background): “Recent work by MacCartney (2009) and MacCartney and Manning (2009) has shown that many new inferential patterns can be derived by tracking how functional expressions project exclusion relations, in addition to the inclusion relations on which monotonicity reasoning is based.”
A Neural Architecture Mimicking Humans End-to-End for Natural Language Inference (2016, background): “NLI is one of the most important component for natural language understanding systems [Benthem, 2008; MacCartney and Manning, 2009].”
From Natural Logic to Natural Reasoning (2015, methodology): “A good place to start is the example (7) in MacCartney and Manning [21]:”
Exploring End-to-End Differentiable Natural Logic Modeling (2020, methodology): “For example, if NLI is deﬁned as a three-way classiﬁcation problem ( entailment , contradiction , and neutral ).”