Generating Logical Forms from Graph Representations of Text and Entities

Shaw, Peter J.; Massey, Philip; Chen, Angelica; Piccinno, Francesco; Altün, Yasemin

doi:10.48550/arxiv.1905.08407

Cited by 4 publications

(10 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A Hidden Markov Model based chart (bar, line, etc) recognition method is proposed in [14]. Graph neural networks have been employed in [5] to generate logical forms with entities from free text using BERT. In a very recent work [6], Obeid et al have used transformer based models for text generation from charts.…”

Section: Related Workmentioning

confidence: 99%

“…Among them a few example tasks are: document summarization [1], title or caption generation from texts, generating textual description of charts [2], named entity recognition [3], etc. There has been several attempts to generate graphs or structural elements from natural language texts or free texts [4,5,6,7]. Scientific charts (bar, line, pie, etc.)…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Text2Chart: A Multi-Staged Chart Generator from Natural Language Text

Mahinur¹,

Kawsar²,

Huzzat³

et al. 2021

Preprint

View full text Add to dashboard Cite

Generation of scientific visualization from analytical natural language text is a challenging task. In this paper, we propose Text2Chart, a multi-staged chart generator method.Text2Chart takes natural language text as input and produce visualization as two-dimensional charts. Text2Chart approaches the problem in three stages. Firstly, it identifies the axis elements of a chart from the given text known as x and y entities. Then it finds a mapping of x-entities with its corresponding y-entities. Next, it generates a chart type suitable for the given text: bar, line or pie. Combination of these three stages is capable of generating visualization from the given analytical text. We have also constructed a dataset for this problem. Experiments show that Text2Chart achieves best performances with BERT based encodings with LSTM models in the first stage to label x and y entities, Random Forest classifier for the mapping stage and fastText embedding with LSTM for the chart type prediction. In our experiments, all the stages show satisfactory results and effectiveness considering formation of charts from analytical text, achieving a commendable overall performance.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Text2Chart: A Multi-Staged Chart Generator from Natural Language Text

Mahinur¹,

Kawsar²,

Huzzat³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…While traditional approaches (Kate & Mooney, 2006;Wong & Mooney, 2007;Clarke et al, 2010;Zettlemoyer & Collins, 2007;Kwiatkowski et al, 2011;Wang et al, 2015;Li et al, 2015;Cai & Yates, 2013;Berant et al, 2013;Quirk et al, 2015;Artzi et al, 2015;Zhang et al, 2017) rely on high-quality lexicons, manually-built templates, and/or domain or representation specific features, in more recent studies neural models with encoder-decoder architectures show impressing results (Dong & Lapata, 2016;Jia & Liang, 2016;Herzig & Berant, 2017;Su & Yan, 2017;B. Chen & Han, 2018;Shaw et al, 2019).…”

Section: Related Workmentioning

confidence: 99%

“…Very recently, Shaw et al (2019) presents an approach that uses a Graph Neural Network (GNN) architecture, which successfully incorporates information about relevant entities and their relations in parsing natural utterances. Similar to Vinyals et al (2015); Jia & Liang (2016); Herzig & Berant (2017), the decoder has a copying mechanism, which can copy an entity to the output during parsing.…”

Section: Related Workmentioning

confidence: 99%

“…With recent advancement of neural networks, neural models with encoder-decoder architectures has obtained impressive improvements (Dong & Lapata, 2016;Jia & Liang, 2016;Herzig & Berant, 2017;Su & Yan, 2017;B. Chen & Han, 2018;Shaw et al, 2019). These encoder-decoder based neural semantic parsers produce one candidate given an input sentence, essentially acting as both the generator and the critic.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Improving Semantic Parsing with Neural Generator-Reranker Architecture

Inan¹,

Tomar²,

Pan³

2019

Preprint

View full text Add to dashboard Cite

Semantic parsing is the problem of deriving machine interpretable meaning representations from natural language utterances. Neural models with encoder-decoder architectures have recently achieved substantial improvements over traditional methods. Although neural semantic parsers appear to have relatively high recall using large beam sizes, there is room for improvement with respect to onebest precision. In this work, we propose a generator-reranker architecture for semantic parsing. The generator produces a list of potential candidates and the reranker, which consists of a pre-processing step for the candidates followed by a novel critic network, reranks these candidates based on the similarity between each candidate and the input sentence. We show the advantages of this approach along with how it improves the parsing performance through extensive analysis. We experiment our model on three semantic parsing datasets (GEO, ATIS, and OVERNIGHT). The overall architecture achieves the state-of-the-art results in all three datasets.

show abstract

Scalable Neural Methods for Reasoning With a Symbolic Knowledge Base

Cohen¹,

Sun²,

Hofer³

et al. 2020

Preprint

View full text Add to dashboard Cite

We describe a novel way of representing a symbolic knowledge base (KB) called a sparse-matrix reified KB. This representation enables neural KB inference modules that are fully differentiable, faithful to the original semantics of the KB, expressive enough to model multi-hop inferences, and scalable enough to use with realistically large KBs. The sparse-matrix reified KB can be distributed across multiple GPUs, can scale to tens of millions of entities and facts, and is orders of magnitude faster than naive sparse-matrix implementations. The reified KB enables very simple end-to-end architectures to obtain competitive performance on several benchmarks representing two families of tasks: KB completion, and learning semantic parsers from denotations.

show abstract

Generating Logical Forms from Graph Representations of Text and Entities

Cited by 4 publications

References 0 publications

Text2Chart: A Multi-Staged Chart Generator from Natural Language Text

Text2Chart: A Multi-Staged Chart Generator from Natural Language Text

Improving Semantic Parsing with Neural Generator-Reranker Architecture

Scalable Neural Methods for Reasoning With a Symbolic Knowledge Base

Contact Info

Product

Resources

About