Learning a Grammar Inducer from Massive Uncurated Instructional Videos
This addresses the problem of learning syntactic grammars from abundant but noisy online instructional videos, which is incremental over prior work on well-aligned data.
The paper tackles video-aided grammar induction with loosely-correlated text-video data from YouTube, achieving higher F1 scores than previous state-of-the-art systems trained on in-domain data across three unseen datasets.
Video-aided grammar induction aims to leverage video information for finding more accurate syntactic grammars for accompanying text. While previous work focuses on building systems for inducing grammars on text that are well-aligned with video content, we investigate the scenario, in which text and video are only in loose correspondence. Such data can be found in abundance online, and the weak correspondence is similar to the indeterminacy problem studied in language acquisition. Furthermore, we build a new model that can better learn video-span correlation without manually designed features adopted by previous work. Experiments show that our model trained only on large-scale YouTube data with no text-video alignment reports strong and robust performances across three unseen datasets, despite domain shift and noisy label issues. Furthermore our model yields higher F1 scores than the previous state-of-the-art systems trained on in-domain data.