Despite the advances in Natural Language Inference through the training of massive deep models, recent work has revealed the generalization difficulties of such models, which fail to perform on adversarial datasets with challenging linguistic phenomena.Such phenomena, however, can be handled well by symbolic systems.Thus, we propose Hy-NLI, a hybrid system that learns to identify an NLI pair as linguistically challenging or not.Based on that, it uses its symbolic or deep learning component, respectively, to make the final inference decision.We show how linguistically less complex cases are best solved by robust state-of-the-art models, like BERT and XLNet, while hard linguistic phenomena are best handled by our implemented symbolic engine.Our thorough evaluation shows that our hybrid system achieves state-of-the-art performance across mainstream and adversarial datasets and opens the way for further research into the hybrid direction.

Deverbal nouns pose serious challenges for knowledge-representation systems. We present a method of canonicalizing deverbal noun representations, relying on a rich lexicon of verb subcategorization frames, the WordNet database, a large finite-state network for derivational morphology and a series of heuristics for mapping deverbal arguments onto the arguments of corresponding verbs.1

We present a stochastic parsing system consisting of a Lexical-Functional Grammar (LFG), a constraint-based parser and a stochastic disambiguation model. We report on the results of applying this system to parsing the UPenn Wall Street Journal (WSJ) treebank. The model combines full and partial parsing techniques to reach full grammar coverage on unseen data. The treebank annotations are used to provide partially labeled data for discriminative statistical estimation using exponential models. Disambiguation performance is evaluated by measuring matches of predicate-argument relations on two distinct test sets. On a gold standard of manually annotated f-structures for a subset of the WSJ treebank, this evaluation reaches 79% F-score. An evaluation on a gold standard of dependency relations for Brown corpus data achieves 76% F-score.

We discuss the treatment of prevention statements in both natural language semantics and knowledge representation, with particular regard to existence entailments. First order representations with an explicit existence predicate are shown to not adequately capture the entailments of prevention statements. A linguistic analysis is framed in a higher order intensional logic, employing a Fregean notion of existence as instantiation of a concept. We discuss how this can be mapped to a Cyc style knowledge representation.

We describe a method for interpreting abstract flat syntactic representations, LFG f-structures, as underspecified semantic representations, here Underspecified Discourse Representation Structures (UDRSs). The method establishes a one-to-one correspondence between subsets of the LFG and UDRS formalisms. It provides a model theoretic interpretation and an inferential component which operates directly on underspecified representations for f-structures through the translation images of f-structures as UDRSs.

Aspects of semantic interpretation, such as quantifier scoping and reference resolution, are often realised computationally by non-monotonic operations involving loss of information and destructive manipulation of semantic representations.The paper describes how monotonic reference resolution and scoping can be carried out using a revised Quasi Logical Form (QLF) representation.Semantics for QLF are presented in which the denotations of formulas are extended monotonically as QLF expressions are resolved.