-
Hi, really brilliant work to accelerate retrieval process in retrieval augumented language models!
I've been working on running through your open-source work so far, but come across some problems …
-
## Summary and motivation
Elasticsearch offers multiple retrieval features including
- approximate dense vector retrieval with embedding inference in Python or in Elasticsearch
- exact dense vect…
-
## Problem statement
1. 적은 양의 (question, passage) pair로 retrieval 성능을 낼 수 있는 training scheme을 찾는다.
2. question과 passage의 내적으로 유사도를 비교할 수 있는 low-dimensional & continuous space에 임베딩한다.
## Baselin…
-
Hello, thanks for your framework.
I'm trying to Training & Evaluating a Dense Retrieval Model on MS MARCO following your instructions in [https://openmatch.readthedocs.io/en/latest/dr-msmarco-passage…
-
CentOS系统Linux，使用conda去装了下openjdk-11-jdk
使用conda install conda-forge/label/gcc7::openjdk去安装openjdk
java -version: openjdk version "11.0.1" 2018-10-16 LTS
OpenJDK Runtime Environment Zulu11.2+3 (buil…
-
After pulling in the most recent change, when I run the following script:
```
from ragatouille import RAGPretrainedModel
def main():
RAG = RAGPretrainedModel.from_pretrained("colbert-ir/…
-
## 一言でいうと
マルチホップのQA(HotpotQAなど)で、再帰的なクエリにより高い精度を記録した研究。Query/PassageそれぞれにEncodeして内積により近さを取り抽出するのが基本だが、QueryのEncode対象に抽出結果をどんどん加えていく。
### 論文リンク
https://arxiv.org/abs/2009.12756
### 著者/所属機関
…
-
Dear Sung-Min Lee,
I sent you an E-mail, but you don't seem to have responded. I'd like to ask you a few more details about the model. I am reaching out to inquire about the data format you used fo…
-
作者你好，最近在阅读《Making Large Language Models A Better Foundation For Dense Retrieval》这篇文章，想复现下论文中pre-task任务，请问对应的代码库的哪一个路径呀？
hahlw updated
5 months ago
-
## 실험 목적
Retrieval하는 방법을 바꿔가며 성능을 비교해보자
## 실험 방법
retrieval_type을 __dense__, __elastic__, __both__(rerank입니다), __sequential__ 로 바꿔가며 실험 진행
(이번 실험은 validation에 있는 240개의 query에 대해서만 성능 분석을 진행…