The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics
arXiv:2102.01672
Abstract
We introduce GEM, a living benchmark for natural language Generation (NLG), its Evaluation, and Metrics. Measuring progress in NLG relies on a constantly evolving ecosystem of automated metrics, datasets, and human evaluation standards. Due to this moving target, new models often still evaluate on divergent anglo-centric corpora with well-established, but flawed, metrics. This disconnect makes it challenging to identify the limitations of current models and opportunities for progress. Addressing this limitation, GEM provides an environment in which models can easily be applied to a wide set of tasks and in which evaluation strategies can be tested. Regular updates to the benchmark will help NLG research become more multilingual and evolve the challenge alongside models. This paper serves as the description of the data for which we are organizing a shared task at our ACL 2021 Workshop and to which we invite the entire NLG community to participate.
References in corpus (15)
- Teaching Machines to Read and Comprehend
- Building Natural-Language Generation Systems
- Multilingual Denoising Pre-training for Neural Machine Translation
- The Natural Language Decathlon: Multitask Learning as Question Answering
- Evaluation of Text Generation: A Survey
- Evaluating the State-of-the-Art of End-to-End Natural Language Generation: The E2E NLG Challenge
- Bringing the People Back In: Contesting Benchmark Machine Learning Datasets
- NUBIA: NeUral Based Interchangeability Assessor for Text Generation
- MultiWOZ 2.2 : A Dialogue Dataset with Additional Annotation Corrections and State Tracking Baselines
- QuestEval: Summarization Asks for Fact-based Evaluation
- GO FIGURE: A Meta Evaluation of Factuality in Summarization
- Beyond Leaderboards: A survey of methods for revealing weaknesses in Natural Language Inference data and models
- AmbigQA: Answering Ambiguous Open-domain Questions
- ORB: An Open Reading Benchmark for Comprehensive Evaluation of Machine Reading Comprehension
- GLGE: A New General Language Generation Evaluation Benchmark
Cited by in corpus (21)
- A Survey of Knowledge-Enhanced Text Generation
- Evaluation of Text Generation: A Survey
- Towards Natural Language Interfaces for Data Visualization: A Survey
- Large Language Models Can Be Strong Differentially Private Learners
- Creativity and Machine Learning: A Survey
- AMMUS : A Survey of Transformer-based Pretrained Models in Natural Language Processing
- Learning by Semantic Similarity Makes Abstractive Summarization Better
- Translation Quality Assessment: A Brief Survey on Manual and Automatic Methods
- Dynaboard: An Evaluation-As-A-Service Platform for Holistic Next-Generation Benchmarking
- Efficient (Soft) Q-Learning for Text Generation with Limited Good Data
- ExplainaBoard: An Explainable Leaderboard for NLP
- Automatic Construction of Evaluation Suites for Natural Language Generation Datasets
- The Benchmark Lottery
- How to Evaluate Your Dialogue Models: A Review of Approaches
- cushLEPOR: customising hLEPOR metric using Optuna for higher agreement with human judgments or pre-trained language model LaBSE
- Benchmarking Multimodal AutoML for Tabular Data with Text Fields
- Human Evaluation of Creative NLG Systems: An Interdisciplinary Survey on Recent Papers
- A Discussion on Building Practical NLP Leaderboards: The Case of Machine Translation
- Datasets: A Community Library for Natural Language Processing
- Truth-Conditional Captioning of Time Series Data
- Compression, Transduction, and Creation: A Unified Framework for Evaluating Natural Language Generation