Pre-training Methods in Information Retrieval

نویسندگان

چکیده

The core of information retrieval (IR) is to identify relevant from large-scale resources and return it as a ranked list respond the user's need. Recently, resurgence deep learning has greatly advanced this field leads hot topic named NeuIR (i.e., neural retrieval), especially paradigm pre-training methods (PTMs). Owing sophisticated objectives huge model size, pre-trained models can learn universal language representations massive textual data, which are beneficial ranking task IR. Since there have been large number works dedicating application PTMs in IR, we believe right time summarize current status, existing researches, gain some insights for future development. In survey, present an overview applied different components IR system, including component, re-ranking other components. addition, also introduce specifically designed available datasets well benchmark leaderboards. Moreover, discuss open challenges envision promising directions, with hope inspiring more on these topics research.

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Analysis of Pre-Processing Methods for Music Information Retrieval in Noisy Environments using Mobile Devices

Recently, content-based music information retrieval (MIR) systems for mobile devices have attracted great interest. However, music retrieval systems are greatly affected by background noise when music is recorded in noisy environments. Therefore, we evaluated various pre-processing methods using the Philips method to determine the one that performs most robust music retrieval in such environmen...

متن کامل

Adaptive data fusion methods in information retrieval

Data fusion is currently used extensively in information retrieval for various tasks. It has proved to be a useful technology because it is able to improve retrieval performance frequently. However, in almost all prior research in data fusion, static search environments have been used, and dynamic search environments have generally not been considered. In this paper, we investigate adaptive dat...

متن کامل

Public Transport Ontology for Passenger Information Retrieval

Passenger information aims at improving the user-friendliness of public transport systems while influencing passenger route choices to satisfy transit user’s travel requirements. The integration of transit information from multiple agencies is a major challenge in implementation of multi-modal passenger information systems. The problem of information sharing is further compounded by the multi-l...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Foundations and Trends in Information Retrieval

سال: 2022

ISSN: ['1554-0669', '1554-0677']

DOI: https://doi.org/10.1561/1500000100