Data Checklist: On Unit-Testing Datasets with Usable Information
Model checklists (Ribeiro et al., 2020) have emerged as a useful tool for understanding the behavior of LLMs, analogous to unit-testing in software engineering. However, despite datasets being a key determinant of model behavior, evaluating datasets, e.g., for the existence of annotation artifacts, is largely done ad hoc, once a problem in model behavior has already been found downstream. In this work, we take a more principled approach to unit-testing datasets by proposing a taxonomy based on the V-information literature. We call a collection of such unit tests a data checklist. Using a checklist, not only are we able to recover known artifacts in well-known datasets such as SNLI, but we also discover previously unknown artifacts in preference datasets for LLM alignment. Data checklists further enable a new kind of data filtering, which we use to improve the efficacy and data efficiency of preference alignment.
Code (1)
Similar Papers 제목 키워드 기반
A Checklist for Deploying Robots in Public: Articulating Tacit Knowledge in the HRI Community
Many of the challenges encountered in in-the-wild public deployments of robots remain undocumented despite sharing many common pitfalls. This creates a high barrier of entry and results in repetition of avoidable mistake…
Beyond Accuracy: Behavioral Testing of NLP models with CheckList
Although measuring held-out accuracy has been the primary approach to evaluate generalization, it often overestimates the performance of NLP models, while alternative approaches for evaluating models either focus on indi…
Question AnsweringSentiment AnalysisWhite-box Testing of NLP models with Mask Neuron Coverage
Recent literature has seen growing interest in using black-box strategies like CheckList for testing the behavior of NLP models. Research on white-box testing has developed a number of methods for evaluating how thorough…
Data AugmentationFault DetectionFine-Grained Fairness Analysis of Abusive Language Detection Systems with CheckList
Current abusive language detection systems have demonstrated unintended bias towards sensitive features such as nationality or gender. This is a crucial issue, which may harm minorities and underrepresented groups if suc…
Abusive LanguageFairnessVL-CheckList: Evaluating Pre-trained Vision-Language Models with Objects, Attributes and Relations
Vision-Language Pretraining (VLP) models have recently successfully facilitated many cross-modal downstream tasks. Most existing works evaluated their systems by comparing the fine-tuned downstream task performance. Howe…