Effective adversary generation for neural machine translation (NMT) is a crucial prerequisite for building robust machine translation systems. In this work, we investigate veritable evaluations of NMT adversarial attacks, and propose a novel method to craft NMT adversarial examples. We first show the current NMT adversarial attacks may be improperly estimated by the commonly used monodirectional translation, and we propose to leverage the round-trip translation technique to build valid metrics for evaluating NMT adversarial attacks. Our intuition is that an effective NMT adversarial example, which imposes minor shifting on the source and degrades the translation dramatically, would naturally lead to a semantic-destroyed round-trip translation result. We then propose a promising black-box attack method called Word Saliency speedup Local Search (WSLS) that could effectively attack the mainstream NMT architectures. Comprehensive experiments demonstrate that the proposed metrics could accurately evaluate the attack effectiveness, and the proposed WSLS could significantly break the state-of-art NMT models with small perturbation. Besides, WSLS exhibits strong transferability on attacking Baidu and Bing online translators.
Seismic anisotropy in the Earth's lower mantle likely results from a combination of elastic anisotropy and lattice preferred orientations of its main constituent minerals. As the second most abundant component of the lower mantle, ferropericlase has been widely studied, and the experimental results demonstrated, in general, a growing with pressure elastic anisotropy up to 1 Mbar. However, the unique measurements on the endmember (MgO) at comparable pressure conditions contradict the above observations and theoretical results. Here, time‐domain Brillouin scattering was applied to measure longitudinal sound velocities in single crystals of MgO compressed in diamond anvil cell. Velocities along two specific crystallographic directions, [100] and [111], were independently collected to 43 GPa. Applying the known bulk modulus, a complete set of single‐crystal elastic moduli, elastic anisotropy and aggregate shear modulus were derived. Our results revealed a steadily increasing with pressure elastic anisotropy at P > 20 GPa, consistent with the previous theoretical predictions and measurements on ferropericlase with moderate amounts of iron.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.