Natural Language Generation in Dialogue using Lexicalized and Delexicalized Data

By Shikhar Sharma, Jing He, Kaheer Suleman, Hannes Schulz, Philip Bachman. 2016

31 citations recorded by the source corpus.

Natural language generation plays a critical role in spoken dialogue systems. We present a new approach to natural language generation for task-oriented dialogue using recurrent neural networks in an encoder-decoder framework. In contrast to previous work, our model uses both lexicalized and delexicalized components i.e. slot-value pairs for dialogue acts, with slots and corresponding values aligned together. This allows our model to learn from all available data including the slot-value pairing, rather than being restricted to delexicalized slots. We show that this helps our model generate more natural sentences with better grammar. We further improve our model's performance by transferring weights learnt from a pretrained sentence auto-encoder. Human evaluation of our best-performing model indicates that it generates sentences which users find more appealing.

Subfields: Artificial Intelligence, 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 Natural Language Processing

Citations in the corpus

Neural Generation for Czech: Data and Baselines (2019, methodology): “Therefore, inspired by Sharma et al. (2017), we test using fully lexicalized input DAs with the main generator to check if it learns to produce more appropriate structure for concrete values (while still producing delexicalized output).”
The E2E Dataset: New Challenges For End-to-End Generation (2017, background): “…development of NLG components in new domains: They jointly learn sentence planning and surface realisation from non-aligned data (Duˇsek and Jurˇc´ıˇcek, 2015; Wen et al., 2015; Mei et al., 2016; Wen et al., 2016; Sharma et al., 2016; Duˇsek and Jurˇc´ıˇcek, 2016a; Lampouras and Vlachos, 2016).”
Why We Need New Evaluation Metrics for NLG (2017, methodology): “, 2015), which bears the danger of ungrammatical lexicalisation (Sharma et al., 2016) and a possible overlap between testing and training set (Lampouras and Vlachos, 2016).”
A Survey of Available Corpora For Building Data-Driven Dialogue Systems: The Journal Version (2015, background): “, 2015), natural language generation (Langkilde and Knight, 1998; Oh and Rudnicky, 2000; Walker et al., 2002; Ratnaparkhi, 2002; Stent et al., 2004; Rieser and Lemon, 2010; Mairesse et al., 2010; Mairesse and Young, 2014; Wen et al., 2015; Sharma et al., 2016), and dialogue policy learning (Young et”
Evaluating the State-of-the-Art of End-to-End Natural Language Generation: The E2E NLG Challenge (2019, background): “2015) as well as other NLG experiments (e.g. Lebret, Grangier, and Auli 2016; Gardent et al. 2017b; Sharma et al. 2016b):”
RankME: Reliable Human Ratings for Natural Language Generation (2018, background): “This is one of the reasons why some more recent research adds a general, overall quality criterion (Wen et al., 2015a,b; Manishina et al., 2016; Novikova et al., 2016, 2017a), or even uses only that (Sharma et al., 2016).”
Relevance of Unsupervised Metrics in Task-Oriented Dialogue for Evaluating Natural Language Generation (2017, methodology): “Research in the task oriented setting has mostly made use of automated metrics such as BLEU and human evaluation (Wen et al., 2015b; Sharma et al., 2017; Dušek and Jurcicek, 2016).”
Findings of the E2E NLG Challenge (2018, background): “…costly semantic alignment between meaning representations (MRs) and the corresponding natural language reference texts, e.g. (Duˇsek and Jurˇc´ıˇcek, 2015; Wen et al., 2015b; Mei et al., 2016; Wen et al., 2016; Sharma et al., 2016; Duˇsek and Jurˇc´ıˇcek, 2016a; Lampouras and Vlachos, 2016).”
Frames: a corpus for adding memory to goal-oriented dialogue systems (2017, background): “To imitate this behaviour, a natural language generator (Oh and Rudnicky, 2000; Wen et al., 2015; Sharma et al., 2017) would need to reason over the database and decide how to tailor the results to the user and present them in a concise but sufficient way.”
Global-to-local Memory Pointer Networks for Task-Oriented Dialogue (2019, neutral): “, 2016), and natural language generation (Sharma et al., 2016) are used.”
Logical Natural Language Generation from Open-Domain Tables (2020, background): “, 2015), which uses rigid slot filling prone to grammatic errors, our fine-grained generation has the flexibility to modify the surface form of non-slot words, which alleviates the linguistic coherence problem (Sharma et al., 2017).”
The aNALoGuE Challenge: Non Aligned Language GEneration (2016, methodology): “This shared task aims to strengthen recent research on corpus-based NLG from unaligned data, e.g. (Duˇsek and Jurcicek, 2015; Wen et al., 2015; Mei et al., 2015; Sharma et al., 2016).”
Semantically Conditioned Dialog Response Generation via Hierarchical Disentangled Self-Attention (2019, background): “Tree Structure First, we note that dialog acts defined in different spoken dialog system (Wen et al., 2016b; Dusek and Jurcı́cek, 2016; Tran et al., 2017; Sharma et al., 2016; Nayak et al., 2017; Budzianowski et al., 2018) have universally unexploited structural property, which is inherently due to ”