Machine Translation Robustness to Natural Asemantic Variation
This addresses robustness issues in machine translation for users dealing with nuanced language inputs, but it is incremental as it builds on existing noise-handling approaches.
The paper tackles the problem of machine translation robustness to natural asemantic variation (NAV), which are minor meaning-preserving input variations, and finds that fine-tuning models with human-generated NAV data improves performance, though synthetic perturbations are less effective.
Current Machine Translation (MT) models still struggle with more challenging input, such as noisy data and tail-end words and phrases. Several works have addressed this robustness issue by identifying specific categories of noise and variation then tuning models to perform better on them. An important yet under-studied category involves minor variations in nuance (non-typos) that preserve meaning w.r.t. the target language. We introduce and formalize this category as Natural Asemantic Variation (NAV) and investigate it in the context of MT robustness. We find that existing MT models fail when presented with NAV data, but we demonstrate strategies to improve performance on NAV by fine-tuning them with human-generated variations. We also show that NAV robustness can be transferred across languages and find that synthetic perturbations can achieve some but not all of the benefits of organic NAV data.