Given an input utterance, we first use a simple method to deterministically generate a set of candidate logical forms with a canonical realization in natural language for each.

Then, we use a paraphrase model to choose the realization that best paraphrases the input, and output the corresponding logical form .

We consider the semantic parsing problem of mapping natural language utterances into logical forms to be executed on a knowledge base (KB) (Zelle and Mooney, 1996; Zettlemoyer and Collins, 2005; Wong and Mooney, 2007; Kwiatkowski et al., 2010).

7igure 1: Semantic parsing via paraphrasing: For each andidate logical form (in red), we generate canonical utter-nces (in purple).

liven an input utterance, we first use a simple de-erministic procedure to construct a manageable et of candidate logical forms (ideally, we would generate canonical utterances for all possible logi-tal forms, but this is intractable).

The trained parser produces a full syntactic parse of any sentence, while simultaneously producing logical forms for portions of the sentence that have a semantic representation within the parser’s predicate vocabulary.

A semantic evaluation demonstrates that this parser produces logical forms better than both comparable prior work and a pipelined syntax-then-semantics approach.

Our parser produces a full syntactic parse of every sentence, and furthermore produces logical forms for portions of the sentence that have a semantic representation within the parser’s predicate vocabulary.

For example, given a phrase like “my favorite town in California,” our parser will assign a logical form like Ax.CITY(x) /\ LOCATEDIN(:E, CALIFORNIA) to the “town in California” portion.

ASP produces a full syntactic analysis of every sentence while simultaneously producing logical forms containing any of 61 category and 69 re-

The input to the parser is a part-of-speech tagged sentence, and the output is a syntactic CCG parse tree, along with zero or more logical forms representing the semantics of subspans of the sentence.

These logical forms are constructed using category and relation predicates from a broad coverage knowledge base.

The parser uses category and relation predicates from a broad coverage knowledge base both to construct logical forms and to parametrize the parsing model.

This line of work has typically used a corpus of sentences with annotated logical forms to train the parser.

We use a CKY algorithm to efficiently determine which phrases the CCG grammar can parse and only allow logical forms for which there exists some context in which they would produce a valid time expression, e.g.

First, we use a CCG to generate an initial logical form for the mention.

initial logical form , as appropriate for its context.

Finally, the logical form is resolved to a TIMEX3 value using a deterministic process.

The MLN constructed to determine the probability of a given entailment includes the logical forms for both T and H as well as soft inference rules that are constructed from distributional information.

To determine an entailment probability, first, the two sentences are mapped to logical representations using Boxer (B08, 2008), a tool for wide-coverage semantic analysis that maps a CCG (Combinatory Categorial Grammar) parse into a lexically-based logical form .

This system uses PSL to compute similarity of logical forms but does not use distributional information on lexical or phrasal similarity.

First, it is explicitly designed to support efficient inference, therefore it scales better to longer sentences with more complex logical forms .

Given the logical forms for a pair of sentences, a text T and a hypothesis H, and given a set of weighted rules derived from the distributional semantics (as explained in section 2.6) composing the knowledge base KB, we build a PSL model that supports determining the truth value of H in the most probable interpretation (i.e.

Parsing into logical form gives:

Both systems map from logical forms with explicit representations of scope of negation out to string-based annotations in the format provided by the Shared Task gold standard.

The main points of difference are in the robustness of the system and in the degree of tailoring of both the rules for determining scope on the logical form level and the rules for handling semantically vacuous elements.

(2012) describe some amount of tailoring of the Boxer lexicon to include more of the Shared Task scope cues among those that produce the negation operator in the DRSs, but otherwise the system appears to directly take the notion of scope of negation from the DRS and project it out to the string, with one caveat: As with the logical-forms representations we use, the DRS logical forms do not include function words as predicates in the semantics.

From these underspecified representations of possible scopal configurations, a scope resolution component can spell out the full range of fully-connected logical forms (Koller and Thater, 2005), but it turns out that such enumeration is not relevant here: the notion of scope encoded in the Shared Task annotations is not concerned with the relative scope of quantifiers and negation, such as the two possible readings of (2) represented informally below:5

Earlier work on concept to text generation mainly focuses on generation from logical forms using rule-based methods.

(Wang, 1980) uses handwritten rules to generate sentences from an extended predicate logic formalism; (Shieber et al., 1990) introduces a head-driven algorithm for generating from logical forms ; (Kay, 1996) defines a chart based algorithm which enhances efficiency by minimising the number of semantically incomplete phrases being built; and (Shemtov, 1996) presents an extension of the chart based generation algorithm presented in (Kay, 1996) which supports the generation of multiple paraphrases from underspecified semantic input.

(Lu and Ng, 2011) focuses on generating natural language sentences from logical form (i.e., lambda terms) using a synchronous context-free grammar.

(2012).3 The dataset includes 880 English questions and their logical forms .

This parser is itself based on SMT, trained on parallel data consisting of English queries and linearized logical forms, and on a language model trained on linearized logical forms .

Recent attempts to learn semantic parsing from question-answer pairs without recurring to annotated logical forms have been presented by Kwiatowski et al.