arXiv (Cornell University) · 2019 · 69 citations · 20 references
ComputingComputer ScienceArtificial IntelligenceMachine LearningNatural Language ProcessingAttention NetworksExplainable Machine LearningLanguage ModelingNatural Language UnderstandingNeural NetworksQuestion AnsweringAttentionAttention MechanismsAttention-based InterpretationInterpretabilityInterpretable Neural NetworksModel InterpretabilityNeural Network InterpretabilityNeural Network Modeling
The attention layer in a neural network model provides insights into the model's reasoning behind its prediction, which are usually criticized for being opaque. Recently, seemingly contradictory viewpoints have emerged about the interpretability of attention weights (Jain & Wallace, 2019; Vig & Belinkov, 2019). Amid such confusion arises the need to understand attention mechanism more systematically. In this work, we attempt to fill this gap by giving a comprehensive explanation which justifies both kinds of observations (i.e., when is attention interpretable and when it is not). Through a series of experiments on diverse NLP tasks, we validate our observations and reinforce our claim of interpretability of attention through manual evaluation.
20
The Measurement of Observer Agreement for Categorical Data
J. Richard Landis, Gary G. Koch · Biometrics · 1977 · 78.9K citations · Full text
Show, Attend and Tell: Neural Image Caption Generation with Visual Attention
Kelvin Xu, Jimmy Ba, Ryan Kiros et al. · arXiv (Cornell University) · 2015 · 7.5K citations · Full text