What Do Position Embeddings Learn? An Empirical Study of Pre-Trained Language Model Positional Encoding
This provides insights for NLP researchers to select appropriate positional encoding functions, though it is incremental as it analyzes existing methods rather than introducing new ones.
The paper investigates what position embeddings learn in pre-trained Transformers and how they affect NLP tasks, finding that they do not always capture positional meaning and that different embeddings impact task performance variably, with specific improvements noted in tasks like machine translation.
In recent years, pre-trained Transformers have dominated the majority of NLP benchmark tasks. Many variants of pre-trained Transformers have kept breaking out, and most focus on designing different pre-training objectives or variants of self-attention. Embedding the position information in the self-attention mechanism is also an indispensable factor in Transformers however is often discussed at will. Therefore, this paper carries out an empirical study on position embeddings of mainstream pre-trained Transformers, which mainly focuses on two questions: 1) Do position embeddings really learn the meaning of positions? 2) How do these different learned position embeddings affect Transformers for NLP tasks? This paper focuses on providing a new insight of pre-trained position embeddings through feature-level analysis and empirical experiments on most of iconic NLP tasks. It is believed that our experimental results can guide the future work to choose the suitable positional encoding function for specific tasks given the application property.