Findings of the E2E NLG Challenge

WS 2018 Ondřej DušekJekaterina NovikovaVerena Rieser

This paper summarises the experimental setup and results of the first shared task on end-to-end (E2E) natural language generation (NLG) in spoken dialogue systems. Recent end-to-end generation systems are promising since they reduce the need for data annotation... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Data-to-Text Generation E2E NLG Challenge TGen BLEU 65.93 # 4
Data-to-Text Generation E2E NLG Challenge TGen NIST 8.6094 # 3
Data-to-Text Generation E2E NLG Challenge TGen METEOR 44.83 # 5
Data-to-Text Generation E2E NLG Challenge TGen ROUGE-L 68.50 # 4
Data-to-Text Generation E2E NLG Challenge TGen CIDEr 2.2338 # 4