Learning Semantic Correspondences with Less Supervision

By Percy Liang, Michael I. Jordan, Daniel Klein. 2009

361 citations recorded by the source corpus.

A central problem in grounded language acquisition is learning the correspondences between a rich world state and a stream of text which references that world state. To deal with the high degree of ambiguity present in this setting, we present a generative model that simultaneously segments the text into utterances and maps each utterance to a meaning representation grounded in the world state. We show that our model generalizes across three domains of increasing difficulty---Robocup sportscasting, weather forecasts (a new domain), and NFL recaps.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

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 Computer Vision and Pattern Recognition

Citations in the corpus

Creating Training Corpora for NLG Micro-Planners (2017, methodology): “generation, the dataset of Liang et al. (2009) includes 29,528 data-text pairs with a vocabulary of 345 words. For the air travel domain, Ratnaparkhi (2000) created a dataset consisting of 5,426 datatext pairs with a richer vocabulary (927 words)”
Joint Concept Learning and Semantic Parsing from Natural Language Explanations (2017, background): “turn, since we never observe the correct interpretation l for any statement, but only model its evaluation zj , we marginalize over all interpretations whose evaluations in a context x matches zj (similar to Liang et al. (2011)).”
Parsing Algebraic Word Problems into Equations (2015, background): “More closely related is work on language grounding, whose goal is the interpretation of a sentence in the context of a world representation (Branavan et al., 2009; Liang et al., 2009; Chen et al., 2010; Bordes et al., 2010; Feng and Lapata, 2010; Hajishirzi et al., 2011; Matuszek et al., 2012; Hajis”
Learning to Solve Arithmetic Word Problems with Verb Categorization (2014, background): “…and Barzi-lay, 2007; Branavan et al., 2009; Branavan et al., 2012; Vogel and Jurafsky, 2010; Chen et al., 2010; Hajishirzi et al., 2011; Chambers and Jurafsky, 2009; Liang et al., 2009; Bordes et al., 2010) where the goal is to align a text into underlying entities and events of an environment.”
DART: Open-Domain Structured Data Record to Text Generation (2020, background): “…Chen and Mooney, 2008; Wiseman et al., 2017; Wang, 2019; Taniguchi et al., 2019), weather forecasts (Goldberg et al., 1994; Reiter et al., 2005; Liang et al., 2009; Konstas and Lapata, 2012; Mei et al., 2016), biographical texts (Lebret et al., 2016; Sha et al., 2018; Liu et al., 2018;…”
Learning Dependency-Based Compositional Semantics (2011, background): “Semantic parsers are typically trained from examples of questions annotated with their target logical forms, but this type of annotation is expensive.”
Weakly Supervised Learning of Semantic Parsers for Mapping Instructions to Actions (2013, methodology): “Data For evaluation, we use the navigation task from MacMahon et al. (2006), which includes three environments and the SAIL corpus of instructions and follower traces.”
Challenges in Data-to-Document Generation (2017, methodology): “In the late 2000s and early 2010s, a number of systems were proposed that did both (Liang et al., 2009; Angeli et al., 2010; Kim and Mooney, 2010; Lu and Ng, 2011; Konstas and Lapata, 2013).”
ToTTo: A Controlled Table-To-Text Generation Dataset (2020, background): “Task Design Most existing table-to-text datasets are restricted in topic and schema such as W EATH - ER G OV (Liang et al., 2009), R OBO C UP (Chen and Mooney, 2008), Rotowire (Wiseman et al., 2017, basketball), E2E (Novikova et al., 2016, 2017, restaurants), KBGen (Banik et al., 2013, biology),…”
Neural Text Generation from Structured Data with Application to the Biography Domain (2016, background): “Hierarchical hidden semi-Markov generative models have also been used to first determine which facts to discuss and then to generate words from the predicates and arguments of the chosen facts (Liang et al., 2009).”
Handling Divergent Reference Texts when Evaluating Table-to-Text Generation (2019, methodology): “In practice, we use simple lexical entailment models to determine this. and Lapata, 2005; Liang et al., 2009).”
Learning to Interpret Natural Language Navigation Instructions from Observations (2011, methodology): “…a semantic parser/language generator to disambiguate the context (Kate and Mooney 2007; Chen, Kim, and Mooney 2010), building a generative model of the content selection process (Liang, Jordan, and Klein 2009; Kim and Mooney 2010), and using a ranking approach (Bor-des, Usunier, and Weston 2010).”
Stochastic variational inference (2012, background): “Many stochastic optimization algorithms beneﬁt from “minibatches,” i.e., several examples at a time (Bottou and Bousquet, 2008; Liang et al., 2009; Mairal et al., 2010).”
What to talk about and how? Selective Generation using LSTMs with Coarse-to-Fine Alignment (2015, methodology): “We attribute this improvement to the RNN’s ability to capture the long-range contextual dependencies that exist between the records, which is known to be essential to selective generation (Barzilay and Lapata, 2005; Liang et al., 2009; Angeli et al., 2010).”
Learning to Win by Reading Manuals in a Monte-Carlo Framework (2011, methodology): “Eisenstein et al. (2009) aim to automatically extract information from a collection of documents to help identify the rules of a game.”
Survey of the State of the Art in Natural Language Generation: Core tasks, applications and evaluation (2017, background): “For example, Liang et al. (2009) extended the work by Barzilay and Lapata (2005) to multiple domains (soccer and weather), relying on weakly supervised techniques; in a similar vein, Koncel-Kedziorski et al. (2014) presented a weakly supervised multilevel approach, to deal with the fact that there…”
Reading to Learn: Constructing Features from Semantic Abstracts (2009, background): “…draws on recent literature on extracting logical forms from surface text (Zettlemoyer and Collins, 2005; Ge and Mooney, 2005; Downey et al., 2005; Liang et al., 2009), interpreting language in the context of a domain (Chen and Mooney, 2008), and using an actionable domain to guide text…”
Training a Multilingual Sportscaster: Using Perceptual Context to Learn Language (2014, methodology): “Recently, Liang et al. (2009) developed a generative model that can be used to match natural-language sentences to facts in a corresponding database to which they may refer.”
Annotating Derivations: A New Evaluation Strategy and Dataset for Algebra Word Problems (2016, methodology): “Prior to the works of Liang et al. (2009) and Clarke et al.”