Compositional Semantic Parsing on Semi-Structured Tables

By Panupong Pasupat, Percy Liang. 2015

1,061 citations recorded by the source corpus.

Two important aspects of semantic parsing for question answering are the breadth of the knowledge source and the depth of logical compositionality. While existing work trades off one aspect for another, this paper simultaneously makes progress on both fronts through a new task: answering complex questions on semi-structured tables using question-answer pairs as supervision. The central challenge arises from two compounding factors: the broader domain results in an open-ended set of relations, and the deeper compositionality results in a combinatorial explosion in the space of logical forms. We propose a logical-form driven parsing algorithm guided by strong typing constraints and show that it obtains significant improvements over natural baselines. For evaluation, we created a new dataset of 22,033 complex questions on Wikipedia tables, which is made publicly available.

Subfields: Artificial Intelligence, Management Science and Operations Research, 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 Management Science and Operations Research

In Natural Language Processing

Citations in the corpus

TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension (2017, background): “, 2014), HTML tables (Pasupat and Liang, 2015), and large (and noisy) corpora (Sawant and Chakrabarti, 2013; Joshi et al.”
DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs (2019, methodology): “It is this literature that we modeled our questions on, particularly looking at the questions in the Wik-iTableQuestions dataset (Pasupat and Liang, 2015).”
CoQA: A Conversational Question Answering Challenge (2018, background): “Another common knowledge source is machine-friendly databases, which organize world facts in the form of a table or a graph (Berant et al., 2013; Pasupat and Liang, 2015; Bordes et al., 2015; Saha et al., 2018; Talmor and Berant, 2018).”
The Next Decade in AI: Four Steps Towards Robust Artificial Intelligence (2020, background): “…working on deriving rich cognitive models from texts (let alone videos) that describe events that develop over time A paper by Pasupat and Liang (Pasupat & Liang, 2015) tries to parse sentences into programmable queries that can be run over tables, but that system doesn't try to accumulate…”
Break It Down: A Question Understanding Benchmark (2020, background): “, 2018), tables (Pasupat and Liang, 2015), and more.”
UL2: Unifying Language Learning Paradigms (2022, methodology): “We use several component tasks from UnifiedSKG (Xie et al., 2022), namely WikiTQ (Pasupat & Liang, 2015), CompWQ (Talmor & Berant, 2018), FetaQA (Nan et al., 2021), HybridQA (Chen et al., 2020), WikiSQL (Zhong et al., 2017), TabFat (Chen et al., 2019), Feverous (Aly et al., 2021), SQA (Iyyer et al.,”
DART: Open-Domain Structured Data Record to Text Generation (2020, background): “As illustrated in Figure 2, DART is constructed using multiple complementary methods: (1) human annotation on open-domain Wikipedia tables from WikiTableQuestions (Pasupat and Liang, 2015) and WikiSQL (Zhong et al., 2017) (Section 3.1), (2) automatic conversion of questions in WikiSQL to declarative”
Seq2SQL: Generating Structured Queries from Natural Language using Reinforcement Learning (2017, background): “WikiTableQuestions [Pasupat and Liang, 2015] is a collection of question and answers, also over a large quantity of tables extracted from Wikipedia.”
UnifiedSKG: Unifying and Multi-Tasking Structured Knowledge Grounding with Text-to-Text Language Models (2022, background): “, 2017) Table Question Answer WikiTQ (Pasupat and Liang, 2015) Table Question Answer CompWebQ (Talmor and Berant, 2018) Knowledge Graph Question Answer HybridQA (Chen et al.”
Axiomatic Attribution for Deep Networks (2017, methodology): “A common approach is to semantically parse the question to its logical form (Liang, 2016; Pasupat & Liang, 2015) using a set of human-authored grammar rules.”
Neural Approaches to Conversational AI (2018, positive): “Answering (SQA) dataset via crowd sourcing by leveraging WikiTableQuestions (WTQ (Pasupat and Liang, 2015)), which contains highly compositional questions associated with HTML tables from Wikipedia.”
The Web as a Knowledge-Base for Answering Complex Questions (2018, methodology): “In future work, we plan to interface with tables (Pasupat and Liang, 2015) and KBs (Zhong et al.”
ToTTo: A Controlled Table-To-Text Generation Dataset (2020, background): “…ability for reasoning and numerical inference (Wiseman et al., 2017) and for building representations of structured objects (Liu et al., 2018), providing an interesting complement to tasks that test these aspects in the NLU setting (Pasupat and Liang, 2015; Chen et al., 2019; Dua et al., 2019).”
Learning to Compose Neural Networks for Question Answering (2016, background): “, 2013) or from (world, question, answer) triples alone (Liang et al., 2011; Pasupat and Liang, 2015).”
Neural Programmer: Inducing Latent Programs with Gradient Descent (2015, background): “Pasupat & Liang (2015) develop a semantic parser with a hand engineered grammar for question answering on tables with natural language questions.”
Neural Symbolic Machines: Learning Semantic Parsers on Freebase with Weak Supervision (2016, background): “This is reminiscent of the floating parser (Wang et al., 2015; Pasupat and Liang, 2015), where a deriva-”
Building a Semantic Parser Overnight (2015, background): “This rather unorthodox approach to semantic parsing can be seen as a generalization of Berant and Liang (2014) and is explained in more detail in Pasupat and Liang (2015).”
Learning Language Games through Interaction (2016, background): “…utter-ance and the logical form, in contrast to most traditional work in semantic parsing (Zettlemoyer and Collins, 2005; Wong and Mooney, 2007; Liang et al., 2011; Kwiatkowski et al., 2010; Berant et al., 2013), instead following a looser model of semantics similar to (Pasupat and Liang, 2015).”
Search-based Neural Structured Learning for Sequential Question Answering (2017, methodology): “One of the main focuses of semantic parsing research is how to address compositionality in language, and complicated questions have been specifically targeted in the design of a recently-released QA dataset (Pasupat and Liang, 2015).”