A Study on Token Pruning for ColBERT
This addresses the storage efficiency issue for users of ColBERT-based retrieval systems, though it is incremental as it builds on existing compression methods.
The study tackled the large index size problem in ColBERT by exploring token pruning techniques, achieving up to 30% reduction in index size on the MS MARCO passage collection without significant performance loss.
The ColBERT model has recently been proposed as an effective BERT based ranker. By adopting a late interaction mechanism, a major advantage of ColBERT is that document representations can be precomputed in advance. However, the big downside of the model is the index size, which scales linearly with the number of tokens in the collection. In this paper, we study various designs for ColBERT models in order to attack this problem. While compression techniques have been explored to reduce the index size, in this paper we study token pruning techniques for ColBERT. We compare simple heuristics, as well as a single layer of attention mechanism to select the tokens to keep at indexing time. Our experiments show that ColBERT indexes can be pruned up to 30\% on the MS MARCO passage collection without a significant drop in performance. Finally, we experiment on MS MARCO documents, which reveal several challenges for such mechanism.