Sean MacAvaney, Franco Maria Nardini, Raffaele Perego, Nicola Tonellotto, Nazli Goharian, and Ophir Frieder. Efficient Document Re-Ranking for Transformers by Precomputing Term Representations. SIGIR 2020.
Code to reproduce the results in the paper is found in OpenNIR.
The PreTTR vocab is a drop-in replacement for a BERT-based vocab by setting: vocab=prettr_bert
.
You can set the join layer with vocab.join_layer=X
(where X
is the layer where cross-attention begins). To enable compression, set vocab.compress_size=X
(where X
is typically 0
(no compression) 384
, 256
, or 128
, but can be anything). To enable FP16, set vocab.compress_fp16=True
.
You can use pre-trained compressions using vocab.bert_weights=X
, where X
is a file name that exists in ~/data/onir/vocab/prettr_bert/
. We have a set of pre-trained compressors from CAR here.
Set these after loading in the model configuration, such as config/vanilla_bert vocab=prettr_bert ...
@inproceedings{macavaney:sigir2020-eff,
author = {MacAvaney, Sean and Nardini, Franco Maria and Perego, Raffaele and Tonellotto, Nicola and Goharian, Nazli and Frieder, Ophir},
title = {Efficient Document Re-Ranking for Transformers by Precomputing Term Representations},
booktitle = {Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval},
year = {2020}
}
Download: Google Drive (21M)
(Tip for downloading large Google Drive files using wget)
File Name | join_layer | compress_size | MD5 |
---|---|---|---|
car.10.128 |
10 | 128 | 1efa63621216c3e2322b2fcec4524917 |
car.10.256 |
10 | 256 | be3be7f8680fa7ffae842acaf2999b71 |
car.10.384 |
10 | 384 | 5eac1cc55d8a9b71c83da8c6c2fbc890 |
car.11.128 |
11 | 128 | 6f87029d7802e64e855a45a081e45068 |
car.11.256 |
11 | 256 | eec1592077c39e93be0c4ef74f0981a0 |
car.11.384 |
11 | 384 | 074e57afef603dfb24b0344f613744d6 |
car.7.128 |
7 | 128 | 02b659407375f634835ab610fbe5db1a |
car.7.256 |
7 | 256 | d2ca951bd09a69db716cf97e24ad606c |
car.7.384 |
7 | 384 | 85fde69e6a6a21e213ee054354343b74 |
car.8.128 |
8 | 128 | b7490664cf042b33dc29988920db9167 |
car.8.256 |
8 | 256 | a8923b7669a6f091d060007e8bff9bf5 |
car.8.384 |
8 | 384 | 2b42a6beb65e76390b762a53140bc774 |
car.9.128 |
9 | 128 | a3ae7780b18a2f00509cc4671530c716 |
car.9.256 |
9 | 256 | 9202a3bea49cf70dd859f6bd0b77dbe8 |
car.9.384 |
9 | 384 | 8b8547e55abe4c12c78c4b946de39adc |