Word Order Matters when you Increase Masking
Word order, an essential property of natural languages, is injected in Transformer-based neural language models using position encoding. However, recent experiments have shown that explicit position encoding is not always useful, since some models without such feature managed to achieve state-of-the art performance on some tasks. To understand better this phenomenon, we examine the effect of removing position encodings on the pre-training objective itself (i.e., masked language modelling), to test whether models can reconstruct position information from co-occurrences alone. We do so by controlling the amount of masked tokens in the input sentence, as a proxy to affect the importance of position information for the task. We find that the necessity of position information increases with the amount of masking, and that masked language models without position encodings are not able to reconstruct this information on the task. These findings point towards a direct relationship between the amount of masking and the ability of Transformers to capture order-sensitive aspects of language using position encoding.
Code (0)
등록된 구현이 없습니다.
Tasks
Language ModellingPositionSentenceMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
When classifying arguments, BERT doesn’t care about word order...except when it matters
When classifying grammatical role, BERT doesn't care about word order... except when it matters
Because meaning can often be inferred from lexical semantics alone, word order is often a redundant cue in natural language. For example, the words cut, chef, and onion are more likely used to convey "The chef cut the on…
Order matters: Distributional properties of speech to young children bootstraps learning of semantic representations
Some researchers claim that language acquisition is critically dependent on experiencing linguistic input in order of increasing complexity. We set out to test this hypothesis using a simple recurrent neural network (SRN…
Language AcquisitionWhen classifying grammatical role, BERT doesn't care about word order... except when it matters
Because meaning can often be inferred from lexical semantics alone, word order is often a redundant cue in natural language. For example, the words chopped, chef, and onion are more likely used to convey "The chef choppe…
When classifying grammatical role, BERT doesn’t care about word order... except when it matters
Because meaning can often be inferred from lexical semantics alone, word order is often a redundant cue in natural language. For example, the words chopped, chef, and onion are more likely used to convey “The chef choppe…