Pruning the Index Contents for Memory Efficient Open-Domain QA
Abstract
This work presents a novel pipeline that demonstrates what is achievable with a combined effort of state-of-the-art approaches. Specifically, it proposes the novel R2-D2 (Rank twice, reaD twice) pipeline composed of retriever, passage reranker, extractive reader, generative reader and a simple way to combine them. Furthermore, previous work often comes with a massive index of external documents that scales in the order of tens of GiB. This work presents a simple approach for pruning the contents of a massive index such that the open-domain QA system altogether with index, OS, and library components fits into 6GiB docker image while retaining only 8% of original index contents and losing only 3% EM accuracy.
- Publication:
-
arXiv e-prints
- Pub Date:
- February 2021
- DOI:
- 10.48550/arXiv.2102.10697
- arXiv:
- arXiv:2102.10697
- Bibcode:
- 2021arXiv210210697F
- Keywords:
-
- Computer Science - Computation and Language;
- Computer Science - Artificial Intelligence;
- Computer Science - Machine Learning
- E-Print:
- v2 - added connection between pruner and DPR, results on TriviaQA, new reranker, results with HN-DPR checkpoint and additional analyses