Evaluating Models' Local Decision Boundaries via Contrast Sets

• URL: http://arxiv.org/abs/2004.02709v2
• Date: Thu, 1 Oct 2020 21:26:57 GMT
• Title: Evaluating Models' Local Decision Boundaries via Contrast Sets
• Authors: Matt Gardner, Yoav Artzi, Victoria Basmova, Jonathan Berant, Ben Bogin, Sihao Chen, Pradeep Dasigi, Dheeru Dua, Yanai Elazar, Ananth Gottumukkala, Nitish Gupta, Hanna Hajishirzi, Gabriel Ilharco, Daniel Khashabi, Kevin Lin, Jiangming Liu, Nelson F. Liu, Phoebe Mulcaire, Qiang Ning, Sameer Singh, Noah A. Smith, Sanjay Subramanian, Reut Tsarfaty, Eric Wallace, Ally Zhang, Ben Zhou
• Abstract summary: We propose a new annotation paradigm for NLP that helps to close systematic gaps in the test data. We demonstrate the efficacy of contrast sets by creating them for 10 diverse NLP datasets. Although our contrast sets are not explicitly adversarial, model performance is significantly lower on them than on the original test sets.
• Score: 119.38387782979474
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Standard test sets for supervised learning evaluate in-distribution generalization. Unfortunately, when a dataset has systematic gaps (e.g., annotation artifacts), these evaluations are misleading: a model can learn simple decision rules that perform well on the test set but do not capture a dataset's intended capabilities. We propose a new annotation paradigm for NLP that helps to close systematic gaps in the test data. In particular, after a dataset is constructed, we recommend that the dataset authors manually perturb the test instances in small but meaningful ways that (typically) change the gold label, creating contrast sets. Contrast sets provide a local view of a model's decision boundary, which can be used to more accurately evaluate a model's true linguistic capabilities. We demonstrate the efficacy of contrast sets by creating them for 10 diverse NLP datasets (e.g., DROP reading comprehension, UD parsing, IMDb sentiment analysis). Although our contrast sets are not explicitly adversarial, model performance is significantly lower on them than on the original test sets---up to 25\% in some cases. We release our contrast sets as new evaluation benchmarks and encourage future dataset construction efforts to follow similar annotation processes.

Related papers

• Evaluating Graph Neural Networks for Link Prediction: Current Pitfalls and New Benchmarking [66.83273589348758]
  Link prediction attempts to predict whether an unseen edge exists based on only a portion of edges of a graph. A flurry of methods have been introduced in recent years that attempt to make use of graph neural networks (GNNs) for this task. New and diverse datasets have also been created to better evaluate the effectiveness of these new models.
  arXiv  Detail & Related papers  (2023-06-18T01:58:59Z)
• Uncertainty Estimation for Language Reward Models [5.33024001730262]
  Language models can learn a range of capabilities from unsupervised training on text corpora. It is often easier for humans to choose between options than to provide labeled data, and prior work has achieved state-of-the-art performance by training a reward model from such preference comparisons. We seek to address these problems via uncertainty estimation, which can improve sample efficiency and robustness using active learning and risk-averse reinforcement learning.
  arXiv  Detail & Related papers  (2022-03-14T20:13:21Z)
• A Closer Look at Debiased Temporal Sentence Grounding in Videos: Dataset, Metric, and Approach [53.727460222955266]
  Temporal Sentence Grounding in Videos (TSGV) aims to ground a natural language sentence in an untrimmed video. Recent studies have found that current benchmark datasets may have obvious moment annotation biases. We introduce a new evaluation metric "dR@n,IoU@m" that discounts the basic recall scores to alleviate the inflating evaluation caused by biased datasets.
  arXiv  Detail & Related papers  (2022-03-10T08:58:18Z)
• Comparing Test Sets with Item Response Theory [53.755064720563]
  We evaluate 29 datasets using predictions from 18 pretrained Transformer models on individual test examples. We find that Quoref, HellaSwag, and MC-TACO are best suited for distinguishing among state-of-the-art models. We also observe span selection task format, which is used for QA datasets like QAMR or SQuAD2.0, is effective in differentiating between strong and weak models.
  arXiv  Detail & Related papers  (2021-06-01T22:33:53Z)
• Automatic Generation of Contrast Sets from Scene Graphs: Probing the Compositional Consistency of GQA [16.95631509102115]
  supervised models often exploit data artifacts to achieve good test scores while their performance severely degrades on samples outside their training distribution. We present a novel method which leverages rich semantic input representation to automatically generate contrast sets for the visual question answering task. We find that, despite GQA's compositionality and carefully balanced label distribution, two high-performing models drop 13-17% in accuracy compared to the original test set.
  arXiv  Detail & Related papers  (2021-03-17T12:19:25Z)
• Learning to Model and Ignore Dataset Bias with Mixed Capacity Ensembles [66.15398165275926]
  We propose a method that can automatically detect and ignore dataset-specific patterns, which we call dataset biases. Our method trains a lower capacity model in an ensemble with a higher capacity model. We show improvement in all settings, including a 10 point gain on the visual question answering dataset.
  arXiv  Detail & Related papers  (2020-11-07T22:20:03Z)
• The Gap on GAP: Tackling the Problem of Differing Data Distributions in Bias-Measuring Datasets [58.53269361115974]
  Diagnostic datasets that can detect biased models are an important prerequisite for bias reduction within natural language processing. undesired patterns in the collected data can make such tests incorrect. We introduce a theoretically grounded method for weighting test samples to cope with such patterns in the test data.
  arXiv  Detail & Related papers  (2020-11-03T16:50:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.