Two-Level Model for Table-to-Text Generation

Cao, Juan; Gong, Jianya; Zhang, Pengzhou

doi:10.1145/3364908.3365287

Cited by 4 publications

(5 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Table 6 compares SAN-T2T with KL (Konstas and Lapata, 2013a), MBW (Mei et al, 2016), and Two-level model (Cao et al, 2019) on WeatherGov. There is no need to apply copy mechanism because its vocabulary size is small and each sample has 36 records of fixed length, but it could also be concluded that beam search has little improvements, which is consistent with Mei et al (2016).…”

Section: Results and Analysismentioning

confidence: 99%

“…As shown in Table 2, SAN-T2T is compared with several previous works, including KN, Template KN (Heafield et al, 2013), NLM, Table NLM (Lebret et al, 2016), Table2Seq-Single (Bao et al, 2019), Structure-aware Seq2Seq (Liu et al, 2018), and Two-level model (Cao, Gong, and Zhang, 2019). Meanwhile, we provide a vanilla Seq2Seq model (without position encoding and content selector compared to SAN-T2T) and a Transformer model (Vaswani et al, 2017) to conduct an ablation study.…”

Section: Experiments and Analysismentioning

confidence: 99%

See 1 more Smart Citation

SAN-T2T: An automated table-to-text generator based on selective attention network

Ding

2023

Nat. Lang. Eng.

View full text Add to dashboard Cite

Table-to-text generation aims to generate descriptions for structured data (i.e., tables) and has been applied in many fields like question-answering systems and search engines. Current approaches mostly use neural language models to learn alignment between output and input based on the attention mechanisms, which are still flawed by the gradual weakening of attention when processing long texts and the inability to utilize the records’ structural information. To solve these problems, we propose a novel generative model SAN-T2T, which consists of a field-content selective encoder and a descriptive decoder, connected with a selective attention network. In the encoding phase, the table’s structure is integrated into its field representation, and a content selector with self-aligned gates is applied to take advantage of the fact that different records can determine each other’s importance. In the decoding phase, the content selector’s semantic information enhances the alignment between description and records, and a featured copy mechanism is applied to solve the rare word problem. Experiments on WikiBio and WeatherGov datasets show that SAN-T2T outperforms the baselines by a large margin, and the content selector indeed improves the model’s performance.

show abstract

Section: Results and Analysismentioning

confidence: 99%

Section: Experiments and Analysismentioning

confidence: 99%

SAN-T2T: An automated table-to-text generator based on selective attention network

Ding

2023

Nat. Lang. Eng.

View full text Add to dashboard Cite

show abstract

“…An unordered record is formed by several key-value pairs, while an entity 𝑘 𝑖 is also composed of several unordered disciplines (Figure 1). proposed a two-level model for table-to-text generation that employs an improved encoding-decoding architecture using two LSTM-RNNs for field prioritization and value prioritization in the encoder segment, and similarly a two-level attention mechanism at the decoding end to obtain the relationship between the words in the text and the fields in the table [5].…”

Section: Formal Description Of the Taskmentioning

confidence: 99%

GCN-based table-to-text generation research

Yang¹,

Yu²

2023

Sixth International Conference on Advanced Electronic Materials, Computers, and Software Engineering (AEMCSE 2023)

View full text Add to dashboard Cite

Table-to-text generation is an important area of text generation, and the process of structured table-to-text generation faces problems such as the "illusion" of over-understanding table data, and the problem of selecting and ordering the content of the generated text from table data. In this paper, we present some important model mechanisms in the history of its development and a new outlook on the future development of table-to-text generation and possible technical routes. A two-stage encoder-decoder approach is analyzed as to why it is superior to its predecessors, and a new outlook is proposed for implementing table-to-text generation tasks based on graph convolutional neural networks.

show abstract

“…This paper is an extended version of our earlier works [4] and [5]. Reference [4] introduced a two-level model for table-to-text generation and [5] improved the model for open domain.…”

Section: Introductionmentioning

confidence: 99%

“…This paper is an extended version of our earlier works [4] and [5]. Reference [4] introduced a two-level model for table-to-text generation and [5] improved the model for open domain. This paper integrates the two with other methods to form a complete and relatively mature architecture, and greatly strengthens the interpretation and analysis.…”

Section: Introductionmentioning

confidence: 99%

Generating Natural Language Descriptions From Tables

Cao

2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

This paper proposes a neural generative architecture, namely NLDT, to generate a natural language short text describing a table which has formal structure and valuable information. Specifically, the architecture maps fields and values of a table to continuous vectors and then generates a natural language description by leveraging the semantics of a table. The NLDT architecture adopts a two-level neural model to make the most of the structure of a table to fully express the relationship between contents. To deal with the problem of out-of-vocabulary, this paper develops a simple and fast word-conversion method that replaces rare words appearing in texts with common field information in tables and directly replicates contents from table to the output sequence according to the field information. Besides, this paper adds the concept of theme to adapt the NLDT architecture to open domain and improves beam search algorithm to strengthen the results in the inference stage. On the WEATHERGOV dataset, the NLDT architecture improves the state-of-the-art BLEU-4 score from 61.01 to 62.89 and the current state-of-the-art F1 score from 73.21 to 78. On the WIKIBIO and WIKITABLE datasets, the NLDT architecture achieves a BLEU-4 score of 45.77 and 38.71 respectively which also outperform the state-of-the-art approaches. Furthermore, this paper introduces a Chinese dataset WIKIBIOCN including 33,244 biographies with corresponding tables. On the WIKIBIOCN dataset, the NLDT architecture achieves a BLEU-4 score of 38.87 and fairly good manual evaluation.

show abstract

Two-Level Model for Table-to-Text Generation

Cited by 4 publications

References 8 publications

SAN-T2T: An automated table-to-text generator based on selective attention network

SAN-T2T: An automated table-to-text generator based on selective attention network

GCN-based table-to-text generation research

Generating Natural Language Descriptions From Tables

Contact Info

Product

Resources

About