Natural Logic for Textual Inference

By Bill MacCartney, Christopher D. Manning. 2007

219 citations recorded by the source corpus.

This paper presents the first use of a computational model of natural logic---a system of logical inference which operates over natural language---for textual inference. Most current approaches to the PASCAL RTE textual inference task achieve robustness by sacrificing semantic precision; while broadly effective, they are easily confounded by ubiquitous inferences involving monotonicity. At the other extreme, systems which rely on first-order logic and theorem proving are precise, but excessively brittle. This work aims at a middle way. Our system finds a low-cost edit sequence which transforms the premise into the hypothesis; learns to classify entailment relations across atomic edits; and composes atomic entailments into a top-level entailment judgment. We provide the first reported results for any system on the FraCaS test suite. We also evaluate on RTE3 data, and show that hybridizing an existing RTE system with our natural logic system yields significant performance gains.

Subfields: Artificial Intelligence, 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 Natural Language Processing

Citations in the corpus

ProoFVer: Natural Logic Theorem Proving for Fact Verification (2021, background): “NatLog (MacCartney and Manning, 2007), building on past theoretical work on natural logic and monotonicity calculus (Van Benthem et al.”
A SICK cure for the evaluation of compositional distributional semantic models (2014, methodology): “Premises and conclusions are simple, lab-made, English sentences involving semantic phenomena frequently ad-3 dressed by formal semanticists, such as generalized quan-tiﬁers (GQs), ellipsis and temporal reference.”
Abductive Commonsense Reasoning (2019, background): “…reasoning draws inspirations from a signiﬁcant body of work on language-based entailment Bowman et al. (2015); Williams et al. (2018b), language-based logic Lakoff (1970); MacCartney & Manning (2007), and language-based common-sense reasoning Mostafazadeh et al. (2016); Zellers et al. (2018).”
Question-Answer Driven Semantic Role Labeling: Using Natural Language to Annotate Natural Language (2015, background): “The idea of expressing the meaning of natural language in terms of natural language is related to natural logic (MacCartney and Manning, 2007), in which they use natural language for logical inference.”
Ordinal Common-sense Inference (2016, background): “While Gi-ampiccolo et al. (2008) extended binary RTE with an “unknown” category, the entailment community has primarily focussed on issues such as paraphrase and monotonicity, such as captured by the Natural Logic implementation of MacCartney and Manning (2007).”
NaturalLI: Natural Logic Inference for Common Sense Reasoning (2014, methodology): “We build upon the variant of the logic introduced by the NatLog system (MacCartney and Manning, 2007; 2008; 2009), based on earlier the-”
Modeling Semantic Containment and Exclusion in Natural Language Inference (2008, methodology): “But in the other ﬁve sections (about 60% of the problems), we achieve accuracy of 87.04%, an error reduction of 61% from (MacCartney and Manning, 2007).”
Differentiable Reasoning on Large Knowledge Bases and Natural Language (2019, background): “A rich and broad literature in MR has approached this problem within a variety of frameworks, including Natural Logic (MacCartney and Manning 2007), Semantic Parsing (Bos 2008), Natural Language Inference and Recognising Textual Entailment (Fyodorov, Winter, and Francez 2000; Bowman et al. 2015), an”
A Monotonicity Calculus and Its Completeness (2017, background): “…Icard, 2012; Moss, 2012; Icard and Moss, 2013; Tune, 2016), and much of this work has made its way into psycholinguists (Geurts, 2003; Geurts and van der Slik, 2005) and natural language processing (MacCartney and Manning, 2007; Angeli and Manning, 2014; Bowman et al., 2015; Abzianidze, 2015).”
Language models show human-like content effects on reasoning (2022, background): “The first task we consider has been studied extensively in the natural language processing literature (MacCartney and Manning, 2007).”
Natural Language Inference with External Knowledge (2017, methodology): “However, this architecture ignores the local interaction between two sentences, which is necessary in traditional natural language inference procedure (MacCartney & Manning, 2007).”
Combining Natural Logic and Shallow Reasoning for Question Answering (2016, methodology): “We use the logic introduced by the NatLog system (MacCartney and Manning, 2007; 2008; 2009), which was in turn based on earlier theoretical work on Monotonicity Calculus (van Benthem, 1986; Sánchez Valencia, 1991).”
Learning semantic relationships for better action retrieval in images (2015, methodology): “Relationship prediction in NLP Our work also draws inspiration from research in NLP such as entailment[23] and natural logic [4].”
Causal Abstractions of Neural Networks (2021, methodology): “This challenging dataset was constructed with a tree-structured natural logic causal model [17, 29, 14].”
Evaluating the Logical Reasoning Ability of ChatGPT and GPT-4 (2023, methodology): “Previous ChatGPT evaluation indicates that ChatGPT performs better on deductive reasoning than inductive ones (Bang et al., 2023) and that ChatGPT performs better on factual inputs (Qin et al., 2023).”