Evaluation in machine learning is usually informed by past choices, for example which datasets or metrics to use. This standardization enables the comparison on equal footing using leaderboards, but the evaluation choices become sub-optimal as better alternatives arise. This problem is especially pertinent in natural language generation which requires ever-improving suites of datasets, metrics, and human evaluation to make definitive claims. To make following best model evaluation practices easier, we introduce GEMv2. The new version of the Generation, Evaluation, and Metrics Benchmark introduces a modular infrastructure for dataset, model, and metric developers to benefit from each others work. GEMv2 supports 40 documented datasets in 51 languages. Models for all datasets can be evaluated online and our interactive data card creation and rendering tools make it easier to add new datasets to the living benchmark.
We introduce Repro, an open-source library which aims at improving the reproducibility and usability of research code. The library provides a lightweight Python API for running software released by researchers within Docker containers which contain the exact required runtime configuration and dependencies for the code. Because the environment setup for each package is handled by Docker, users do not have to do any configuration themselves. Once Repro is installed, users can run the code for the 30+ papers currently supported by the library. We hope researchers see the value provided to others by including their research code in Repro and consider adding support for their own research code.
How reliably an automatic summarization evaluation metric replicates human judgments of summary quality is quantified by system-level correlations. We identify two ways in which the definition of the system-level correlation is inconsistent with how metrics are used to evaluate systems in practice and propose changes to rectify this disconnect. First, we calculate the system score for an automatic metric using the full test set instead of the subset of summaries judged by humans, which is currently standard practice. We demonstrate how this small change leads to more precise estimates of system-level correlations. Second, we propose to calculate correlations only on pairs of systems that are separated by small differences in automatic scores which are commonly observed in practice. This allows us to demonstrate that our best estimate of the correlation of ROUGE to human judgments is near 0 in realistic scenarios. The results from the analyses point to the need to collect more high-quality human judgments and to improve automatic metrics when differences in system scores are small.
Question answering-based summarization evaluation metrics must automatically determine whether the QA model's prediction is correct or not, a task known as answer verification. In this work, we benchmark the lexical answer verification methods which have been used by current QA-based metrics as well as two more sophisticated text comparison methods, BERTScore and LERC. We find that LERC out-performs the other methods in some settings while remaining statistically indistinguishable from lexical overlap in others. However, our experiments reveal that improved verification performance does not necessarily translate to overall QA-based metric quality: In some scenarios, using a worse verification method -- or using none at all -- has comparable performance to using the best verification method, a result that we attribute to properties of the datasets.
In this work, we propose a method for incorporating question-answering (QA) signals into a summarization model. Our method identifies salient noun phrases (NPs) in the input document by automatically generating wh-questions that are answered by the NPs and automatically determining whether those questions are answered in the gold summaries. This QA-based signal is incorporated into a two-stage summarization model which first marks salient NPs in the input document using a classification model, then conditionally generates a summary. Our experiments demonstrate that the models trained using QA-based supervision generate higher-quality summaries than baseline methods of identifying salient spans on benchmark summarization datasets. Further, we show that the content of the generated summaries can be controlled based on which NPs are marked in the input document. Finally, we propose a method of augmenting the training data so the gold summaries are more consistent with the marked input spans used during training and show how this results in models which learn to better exclude unmarked document content.
The quality of a summarization evaluation metric is quantified by calculating the correlation between its scores and human annotations across a large number of summaries. Currently, it is not clear how precise these correlation estimates are, nor whether differences between two metrics' correlations reflects a true difference or if it is due to random chance. In this work, we address these two problems by proposing methods for calculating confidence intervals and running hypothesis tests for correlations using two resampling methods, bootstrapping and permutation. After evaluating which of the proposed methods is most appropriate for summarization through two simulation experiments, we analyze the results of applying these methods to several different automatic evaluation metrics across three sets of human annotations. We find that the confidence intervals are rather wide, demonstrating high uncertainty in how reliable automatic metrics truly are. Further, although many metrics fail to show statistical improvements over ROUGE, two recent works, QAEval and BERTScore, do in some evaluation settings.
Reference-based metrics such as ROUGE or BERTScore evaluate the content quality of a summary by comparing the summary to a reference. Ideally, this comparison should measure the summary's information quality by calculating how much information the summaries have in common. In this work, we analyze the token alignments used by ROUGE and BERTScore to compare summaries and argue that their scores largely cannot be interpreted as measuring information overlap, but rather the extent to which they discuss the same topics. Further, we provide evidence that this result holds true for many other summarization evaluation metrics. The consequence of this result is that it means the summarization community has not yet found a reliable automatic metric that aligns with its research goal, to generate summaries with high-quality information. Then, we propose a simple and interpretable method of evaluating summaries which does directly measure information overlap and demonstrate how it can be used to gain insights into model behavior that could not be provided by other methods alone.
Recently, there has been growing interest in using question-answering (QA) models to evaluate the content quality of summaries. While previous work has shown initial promising results in this direction, their experimentation has been limited, leading to a poor understanding of the utility of QA in evaluating summary content. In this work, we perform an extensive evaluation of a QA-based metric for summary content quality, calculating its performance with today's state-of-the-art models as well as estimating its potential upper-bound performance. We analyze a proposed metric, QAEval, which is more widely applicable than previous work. We show that QAEval already achieves state-of-the-art performance at scoring summarization systems, beating all other metrics including the gold-standard Pyramid Method, while its performance on individual summaries is at best competitive to other automatic metrics. Through a careful analysis of each component of QAEval, we identify the performance bottlenecks and estimate that with human-level performance, QAEval's summary-level results have the potential to approach that of the Pyramid Method.
We present SacreROUGE, an open-source library for using and developing summarization evaluation metrics. SacreROUGE removes many obstacles that researchers face when using or developing metrics: (1) The library provides Python wrappers around the official implementations of existing evaluation metrics so they share a common, easy-to-use interface; (2) it provides functionality to evaluate how well any metric implemented in the library correlates to human-annotated judgments, so no additional code needs to be written for a new evaluation metric; and (3) it includes scripts for loading datasets that contain human judgments so they can easily be used for evaluation. This work describes the design of the library, including the core Metric interface, the command-line API for evaluating summarization models and metrics, and the scripts to load and reformat publicly available datasets. The development of SacreROUGE is ongoing and open to contributions from the community.