- Home
- People
- Slugbot
- Publications
- NLDS Corpora for Downloads
- Internet Argument Corpus
- Internet Argument Corpus v2
- Argument Extraction Corpus
- Argumentative Dialogue Summary Corpus
- Fact-Feeling Dataset
- Sarcasm Corpus V1
- Sarcasm Corpus V2
- Sarcasm Rhetorical Questions Corpus
- Film Corpus 1.0
- Film Corpus 2.0
- Argument Facet Similarity Corpus
- First-Person Sentiment Corpus
- Hotel Corpora
- PersonaBank
- PersonageNLG: Style in NLG
- Sentence Planning Corpus for NLG
- SlugNERDS: NER Resources
- Source Blending in NLG
- The Persuasion and Personality Corpus
- ViGGO: NLG for Video Game Conversations
- YelpNLG: Review Corpus for NLG
- Contingent Causal Event Pairs
- Contingent Pairs of Fine-Grained Everyday Events
- Narrative Causality Event Pairs
- DesireDB
- Story Dialogue with Gestures (SDG) Corpus
- Artwalk Corpus
- Storytron Retellings Corpus
- Personality Dyads Corpus
Source Blending in NLG
L. Reed, V.Harrison, S. Oraby, D. Hakkani-Tür, and M. Walker. SIGDIAL 2020. Boise, Idaho.
Overview: The Source Blending NLG corpus is a set of 77k meaning representation to natural language utterance pairs in the restaurant domain from two source ontologies, NYC and E2E (train) and 3040 MRs with attributes from both sources (test).
Data: The data available for download is a zip of 2 CSVs files (one for train, one for test) containing meaning representations and their corresponding natural language utterances (utterances only for train)
Download: Fill out the following form to download the source blending for NLG corpus.