Robustness Gym: Unifying the NLP Evaluation Landscape

By Karan Goel, Nazneen Rajani, Jesse Vig, Samson Tan, Jason M. Wu, Stephan Zheng, Caiming Xiong, Mohit Bansal, Christopher R'e. 2021

167 citations recorded by the source corpus.

Despite impressive performance on standard benchmarks, natural language processing (NLP) models are often brittle when deployed in real-world systems. In this work, we identify challenges with evaluating NLP systems and propose a solution in the form of Robustness Gym (RG), a simple and extensible evaluation toolkit that unifies 4 standard evaluation paradigms: subpopulations, transformations, evaluation sets, and adversarial attacks. By providing a common platform for evaluation, RG enables practitioners to compare results from disparate evaluation paradigms with a single click, and to easily develop and share novel evaluation methods using a built-in set of abstractions. RG is under active development and we welcome feedback & contributions from the community.

Subfields: Artificial Intelligence, Evaluation & Benchmarks

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 Evaluation & Benchmarks

Citations in the corpus

Repairing the Cracked Foundation: A Survey of Obstacles in Evaluation Practices for Generated Text (2022, background): “schemes more robust, Gorman and Bedrick (2019) propose using multiple random splits similar to cross-validation as they find that results on multiple NLU tasks change when the splitting process is changed.”
Datasets: A Community Library for Natural Language Processing (2021, methodology): “The Robustness Gym 9 proposes a systematic way to test an NLP system across many different proposed techniques, speciﬁ-cally subpopulations, transformations, evaluation sets, and adversarial attacks (Goel et al., 2021).”
An Empirical Survey of Data Augmentation for Limited Data Learning in NLP (2021, background): “For more detailed discussion on textual adversarial examples, please refer to recent comprehensive surveys (Zhang et al., 2019; Huq and Pervin, 2020; Goel et al., 2021).”
SummVis: Interactive Visual Analysis of Models, Data, and Evaluation for Text Summarization (2021, methodology): “Several recent works have realized the need for evolving benchmarks and evaluations (Goel et al., 2021; Gehrmann et al., 2021; Khashabi et al., 2021).”
Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models (2021, methodology): “( i ) Evaluation toolkits, including OpenAttack [58], TextAttack [35], TextFlint [17] and Robustness Gym [15], integrate various ad hoc input transformations for different tasks and provide programmable APIs to dynamically test model performance.”
Dynaboard: An Evaluation-As-A-Service Platform for Holistic Next-Generation Benchmarking (2021, background): “• Explainaboard [39] and Robustness Gym [23] do allow for a more nuanced comparison of models, but largely through the lens of which slices of data a model performs well on and what types of errors it makes.”
Interactive Model Cards: A Human-Centered Approach to Model Documentation (2022, methodology): “Robustness Gym [23] and Gensim [58] are primarily used to handle data inputs and outputs between the cards visual components, and are supported by other packages (nltk [9], numpy [27], pandas [42], sklearn [53]).”
NL-Augmenter: A Framework for Task-Sensitive Natural Language Augmentation (2021, methodology): “In this paper, we present a participant-driven repository for creating and testing transformations and filters , and for applying them to all dataset splits (training, development, evaluation) and to all NLP tasks (NLG, labeling, question answering, etc.).”
SEAL : Interactive Tool for Systematic Error Analysis and Labeling (2022, background): “Model evaluation should ideally give actionable insights into a model’s performance on a dataset in the form of data curation (Liang and Zou, 2022) or model patching (Goel et al., 2021a).”
Auditing large language models: a three-layered approach (2023, methodology): “Therefore, researchers and developers have created tools and methods to assess LLMs’ robustness, including evaluation toolkits like the Robustness Gym [192], benchmark datasets like ANLI [193], and open-source platforms like Dynabench [194].”