Modularized Transfomer-based Ranking Framework
This work addresses efficiency and transparency issues in information retrieval for users needing faster and more understandable ranking systems, though it is incremental as it builds on existing Transformer architectures.
The paper tackled the problem of computational expense and lack of interpretability in Transformer-based ranking models by modularizing them into separate text representation and interaction modules, resulting in substantially faster ranking with offline pre-computed representations and improved interpretability.
Recent innovations in Transformer-based ranking models have advanced the state-of-the-art in information retrieval. However, these Transformers are computationally expensive, and their opaque hidden states make it hard to understand the ranking process. In this work, we modularize the Transformer ranker into separate modules for text representation and interaction. We show how this design enables substantially faster ranking using offline pre-computed representations and light-weight online interactions. The modular design is also easier to interpret and sheds light on the ranking process in Transformer rankers.