Conformer-Kernel with Query Term Independence for Document Retrieval
arXiv:2007.10434
Abstract
The Transformer-Kernel (TK) model has demonstrated strong reranking performance on the TREC Deep Learning benchmark---and can be considered to be an efficient (but slightly less effective) alternative to BERT-based ranking models. In this work, we extend the TK architecture to the full retrieval setting by incorporating the query term independence assumption. Furthermore, to reduce the memory complexity of the Transformer layers with respect to the input sequence length, we propose a new Conformer layer. We show that the Conformer's GPU memory requirement scales linearly with input sequence length, making it a more viable option when ranking long documents. Finally, we demonstrate that incorporating explicit term matching signal into the model can be particularly useful in the full retrieval setting. We present preliminary results from our work in this paper.
References in corpus (8)
- Linformer: Self-Attention with Linear Complexity
- Generating Long Sequences with Sparse Transformers
- Deeper Text Understanding for IR with Contextual Neural Language Modeling
- Reformer: The Efficient Transformer
- Pre-training Tasks for Embedding-based Large-scale Retrieval
- Sparse Sinkhorn Attention
- An Updated Duet Model for Passage Re-ranking
- Reply With: Proactive Recommendation of Email Attachments