Assessing the Benchmarking Capacity of Machine Reading Comprehension Datasets
This work addresses the need for more precise benchmarking in machine reading comprehension, which is crucial for researchers and practitioners developing and evaluating language understanding systems, though it is incremental as it builds on existing dataset analysis.
The authors tackled the problem of evaluating the benchmarking capacity of machine reading comprehension datasets by proposing a semi-automated, ablation-based methodology, finding that on average, baseline model scores remained high (e.g., 89.2% and 78.5% of original) when key features were removed, indicating many questions do not require intended skills like grammatical reasoning.
Existing analysis work in machine reading comprehension (MRC) is largely concerned with evaluating the capabilities of systems. However, the capabilities of datasets are not assessed for benchmarking language understanding precisely. We propose a semi-automated, ablation-based methodology for this challenge; By checking whether questions can be solved even after removing features associated with a skill requisite for language understanding, we evaluate to what degree the questions do not require the skill. Experiments on 10 datasets (e.g., CoQA, SQuAD v2.0, and RACE) with a strong baseline model show that, for example, the relative scores of a baseline model provided with content words only and with shuffled sentence words in the context are on average 89.2% and 78.5% of the original score, respectively. These results suggest that most of the questions already answered correctly by the model do not necessarily require grammatical and complex reasoning. For precise benchmarking, MRC datasets will need to take extra care in their design to ensure that questions can correctly evaluate the intended skills.