探索空白空间：人在回路的数据增强
Exploring Empty Spaces: Human-in-the-Loop Data Augmentation
AI总结：
本文提出交互式工具Amplio，通过人在回路的数据增强方法系统识别文本数据中的空白空间，帮助红队成员高效生成多样化、高质量的安全测试提示。
摘要：
数据增强对于使机器学习模型更加稳健和安全至关重要。然而，数据增强可能具有挑战性，因为它需要生成多样化的数据点，以严格评估模型在边缘情况下的行为并减轻潜在危害。创建覆盖这些“未知的未知”的高质量增强数据是一项耗时且需要创造力的任务。在这项工作中，我们介绍了Amplio，这是一种交互式工具，旨在帮助从业者在非结构化文本数据集中探索“未知的未知”，并通过系统性地识别待探索的空白数据空间来提高数据多样性。Amplio包含三种人在回路的数据增强技术：基于概念增强（Augment With Concepts）、通过插值增强（Augment by Interpolation）以及使用大语言模型增强（Augment with Large Language Model）。在一项有18名专业红队成员参与的用户研究中，我们证明了我们的增强方法在帮助生成高质量、多样化且相关的模型安全提示方面的实用性。我们发现Amplio使红队成员能够快速且富有创造性地增强数据，突显了交互式增强工作流的变革潜力。
英文摘要：
Data augmentation is crucial to make machine learning models more robust and safe. However, augmenting data can be challenging as it requires generating diverse data points to rigorously evaluate model behavior on edge cases and mitigate potential harms. Creating high-quality augmentations that cover these "unknown unknowns" is a time- and creativity-intensive task. In this work, we introduce Amplio, an interactive tool to help practitioners navigate "unknown unknowns" in unstructured text datasets and improve data diversity by systematically identifying empty data spaces to explore. Amplio includes three human-in-the-loop data augmentation techniques: Augment With Concepts, Augment by Interpolation, and Augment with Large Language Model. In a user study with 18 professional red teamers, we demonstrate the utility of our augmentation methods in helping generate high-quality, diverse, and relevant model safety prompts. We find that Amplio enabled red teamers to augment data quickly and creatively, highlighting the transformative potential of interactive augmentation workflows.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
