Release date: May 9, 2022
- Added BEIR (v1.0.0) regressions for "flat" baseline, "multfield" baseline, and SPLADE-distill CoCodenser-medium.
- Added BEIR (V1.0.0) qrel and topic bindings.
- Added Rocchio feedback reranker.
- Added multi-threaded method to fetch raw documents from index in batch.
- Added Farsi analyzer.
- Added support for indexing AfriBERTa corpus.
- Refactored MS MARCO v1 and v2 uniCOIL segmented doc regressions (prepending title to segment text during encoding)
Sorted by number of commits:
- Jimmy Lin (lintool)
- Matt Yang (justram)
- Yuqi Liu (yuki617)
- Nandan Thakur (NThakur20)
- Areel Ullah Khan (AreelKhan)
- Xinyu (Crystina) Zhang (crystina-z)
- Habeeb Shopeju (HAKSOAT)
- Jasper Xian (jasper-xian)
- Noam Cohen (noam1023)
- Ogundepo Odunayo (ToluClassics)
- Ji Xi Yang (jx3yang)
- Alvin Dai (alvind1)
All contributors with five or more commits, sorted by number of commits, according to GitHub:
- Jimmy Lin (lintool)
- Peilin Yang (Peilin-Yang)
- Ahmet Arslan (iorixxx)
- Edwin Zhang (edwinzhng)
- Rodrigo Nogueira (rodrigonogueira4)
- Xueguang Ma (MXueguang)
- Emily Wang (emmileaf)
- Royal Sequiera (rosequ)
- Chris Kamphuis (Chriskamphuis)
- Victor Yang (Victor0118)
- Boris Lin (borislin)
- Tommaso Teofili (tteofili)
- Yuqi Liu (yuki617)
- Matt Yang (justram)
- Nikhil Gupta (nikhilro)
- Shane Ding (shaneding)
- Yuhao Xie (Kytabyte)
- Stephanie Hu (stephaniewhoo)
- Kuang Lu (lukuang)
- Ronak Pradeep (ronakice)
- Adam Yang (adamyy)
- Xinyu Mavis Liu (x389liu)
- Luchen Tan (LuchenTan)
- Joel Mackenzie (JMMackenzie)
- Salman Mohammed (salman1993)
- Johnson Han (x65han)
- Zhiying Jiang (bazingagin)
- Matt Yang (d1shs0ap)
- Kelvin Jiang (kelvin-jiang)
- Hang Cui (HangCui0510)
- Michael Tu (tuzhucheng)
- Dayang Shi (dyshi)
- Nandan Thakur (NThakur20)
- Xinyu (Crystina) Zhang (crystina-z)
- Ryan Clancy (ryan-clancy)
- Peng Shi (Impavidity)
- Zeynep Akkalyoncu Yilmaz (zeynepakkalyoncu)