Learning for Semantic Parsing with Statistical Machine Translation
By Yuk Wah Wong, Raymond J. Mooney. 2006
322 citations recorded by the source corpus.
We present a novel statistical approach to semantic parsing, WASP, for constructing a complete, formal meaning representation of a sentence. A semantic parser is learned given a set of sentences annotated with their correct meaning representations. The main innovation of WASP is its use of state-of-the-art statistical machine translation techniques. A word alignment model is used for lexical acquisition, and the parsing model itself can be seen as a syntax-based translation model. We show that WASP performs favorably in terms of both accuracy and coverage compared to existing learning methods requiring similar amount of supervision, and shows better robustness to variations in task complexity and word order.
Subfields: Artificial Intelligence, Machine Translation, Information Extraction
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
- Luke S. Zettlemoyer
- Mohit Bansal
- Percy Liang
- Jianfeng Gao
- Noah A. Smith
- Graham Neubig
- Yejin Choi
- Christopher D. Manning
In Machine Translation
- Jayant Krishnamurthy
- Luke S. Zettlemoyer
- Mirella Lapata
- Raymond J. Mooney
- Sharon Goldwater
- Percy Liang
- Daniel Klein
- Jacob Andreas
In Information Extraction
- Daniel Klein
- Percy Liang
- Raymond J. Mooney
- Luke S. Zettlemoyer
- Christopher D. Manning
- Jayant Krishnamurthy
- Dan Roth
- Jacob Andreas
Citations in the corpus
- Show Your Work: Scratchpads for Intermediate Computation with Language Models (2021, neutral): “Finally, supervised approaches to semantic parsing (Zelle & Mooney, 1996; Zettlemoyer & Collins, 2005; Kwiatkowksi et al., 2010; Wong & Mooney, 2006) predict the text of a database query, which can then be executed to answer a natural language question.”
- Learning to Automatically Solve Algebra Word Problems (2014, background): “Other approaches have relied on access to more costly annotated logical forms (Zelle and Mooney, 1996; Thompson and Mooney, 2003; Wong and Mooney, 2006; Zettlemoyer and Collins, 2005; Kwiatkowski et al., 2010).”
- Learning Dependency-Based Compositional Semantics (2011, background): “• Wasp (Wong and Mooney, 2006) uses log-linear synchronous grammars to transform utterances into logical forms, starting with word alignments obtained from the IBM models.”
- Latent Predictor Networks for Code Generation (2016, background): “Our problem can also be viewed in the framework of semantic parsing (Wong and Mooney, 2006; Lu et al., 2008; Jones et al., 2012; Artzi et al., 2015).”
- Data Recombination for Neural Semantic Parsing (2016, methodology): “W ASP also requires the implementation of an efficient algorithm for computing the conditional probability p ( y | x ) .”
- Modeling Biological Processes for Reading Comprehension (2014, background): “In recent years, several lines of work addressed semantic parsing using various formalisms and levels of supervision (Zettlemoyer and Collins, 2005; Wong and Mooney, 2006; Clarke et al., 2010; Berant et al., 2013).”
- Jointly Learning to Parse and Perceive: Connecting Natural Language to the Physical World (2013, background): “Other variants of grounded language acquisition eliminate the need for perception by assuming access to a logical representation of the environment (Zettlemoyer and Collins, 2005; Wong and Mooney, 2006; Matuszek et al., 2010; Chen and Mooney, 2011; Liang et al., 2011).”
- Language to Logical Form with Neural Attention (2016, methodology): “The phrasal method uses phrase-based machine translation to generate the recipe, whereas sync extracts synchronous grammar rules from the data, essentially recreating WASP (Wong and Mooney, 2006).”
- Training a Multilingual Sportscaster: Using Perceptual Context to Learn Language (2014, methodology): “Wong and Mooney (2006) adapted such methods to learn to translate from NL to MRL rather than from one NL to another.”
- Neural Semantic Parsing with Type Constraints for Semi-Structured Tables (2017, background): “…formalisms such as Combinatory Categorial Grammar (Zettlemoyer 1516 and Collins, 2005, 2007; Kwiatkowski et al., 2011, 2013; Krishnamurthy and Mitchell, 2012; Artzi and Zettlemoyer, 2013) and others (Liang et al., 2011; Berant et al., 2013; Zhao and Huang, 2015; Wong and Mooney, 2006, 2007).”
- Alignment-Based Compositional Semantics for Instruction Following (2015, methodology): “While alignment models from machine translation have previously been used to align words to fragments of semantic parses (Wong and Mooney, 2006; Pourdamghani et al., 2014), we are unaware of such models being used to align entire instruction sequences to demonstrations.”
- Language to Code: Learning Semantic Parsers for If-This-Then-That Recipes (2015, methodology): “Unlike WASP whose synchronous grammar needs to be able to directly parse the input, KRISP’s approach to “soft matching” productions allows it to produce a parse for any input sentence.”
- Compositional Generalization and Natural Language Variation: Can a Semantic Parsing Approach Handle Both? (2020, methodology): “Unlike traditional SCFG models for machine translation applied to semantic parsing (Wong and Mooney, 2006; Andreas et al., 2013), our neural model conditions on global context from the source x via contextual word embeddings, and our grammar’s rules do not need to carry source context to aid disambi”
- Lexical Generalization in CCG Grammar Induction for Semantic Parsing (2011, methodology): “…for learning semantic parsers, including ones based on machine translation techniques (Papineni et al., 1997; Ramaswamy and Kleindienst, 2000; Wong and Mooney, 2006), parsing models (Miller et al., 1996; Ge and Mooney, 2006; Lu et al., 2008), inductive logic programming algorithms (Zelle and…”
- Semantic Parsing with Bayesian Tree Transducers (2012, methodology): “Like the hybrid tree semantic parser (Lu et al., 2008) and the synchronous grammar based WASP (Wong and Mooney, 2006), our model simultaneously generates the input MR tree and the output NL string.”
- Data-Driven Response Generation in Social Media (2011, methodology): “…been applied to a sm¨org˚asbord of NLP problems, including question answering (Echihabi and Marcu, 2003), semantic parsing and generation (Wong and Mooney, 2006; Wong and Mooney, 2007), summarization (Daum´e III and Marcu, 2009), generating bid-phrases in on-line advertising (Ravi et al.,…”
- Abstract Meaning Representation for Sembanking (2013, background): “Researchers working on whole-sentence semantic parsing today typically use small, domain-specific sembanks like GeoQuery (Wong and Mooney, 2006).”
- Online Learning of Relaxed CCG Grammars for Parsing to Logical Form (2007, methodology): “of techniques have been considered including approaches based on machine translation techniques (Papineni et al., 1997; Ramaswamy and Kleindienst, 2000; Wong and Mooney, 2006), parsing techniques (Miller et al.”