Word Shape Matters: Robust Machine Translation with Visual Embedding
This addresses robustness issues in machine translation for users dealing with noisy text, but it is incremental as it builds on existing character-level models.
The paper tackles the problem of neural machine translation models failing on misspelled or substandard inputs by introducing visual embeddings that encode character shapes from printed letter images, resulting in improved robustness even against unseen noises during testing.
Neural machine translation has achieved remarkable empirical performance over standard benchmark datasets, yet recent evidence suggests that the models can still fail easily dealing with substandard inputs such as misspelled words, To overcome this issue, we introduce a new encoding heuristic of the input symbols for character-level NLP models: it encodes the shape of each character through the images depicting the letters when printed. We name this new strategy visual embedding and it is expected to improve the robustness of NLP models because humans also process the corpus visually through printed letters, instead of machinery one-hot vectors. Empirically, our method improves models' robustness against substandard inputs, even in the test scenario where the models are tested with the noises that are beyond what is available during the training phase.