FastFusionNet: New State-of-the-Art for DAWNBench SQuAD

February 28, 2019 · Entered Twilight · 🏛 arXiv.org

"Last commit was 6.0 years ago (≥5 year threshold)"

Evidence collected by the PWNC Scanner

Repo contents: .gitignore, LICENSE, README.md, download.sh, eval.py, prepro.py, qa, requirements.txt, train.py

Authors Felix Wu, Boyi Li, Lequn Wang, Ni Lao, John Blitzer, Kilian Q. Weinberger arXiv ID 1902.11291 Category cs.CL: Computation & Language Citations 5 Venue arXiv.org Repository https://github.com/felixgwu/FastFusionNet ⭐ 39 Last Checked 2 months ago

Abstract

In this technical report, we introduce FastFusionNet, an efficient variant of FusionNet [12]. FusionNet is a high performing reading comprehension architecture, which was designed primarily for maximum retrieval accuracy with less regard towards computational requirements. For FastFusionNets we remove the expensive CoVe layers [21] and substitute the BiLSTMs with far more efficient SRU layers [19]. The resulting architecture obtains state-of-the-art results on DAWNBench [5] while achieving the lowest training and inference time on SQuAD [25] to-date. The code is available at https://github.com/felixgwu/FastFusionNet.