DeepMTL2R: A Library for Deep Multi-task Learning to Rank Conference: Make sure to enter the correct conference title from your rights confirmation emai; June 03–05, 2018; Woodstock, NY ISBN: 978-1-4503-XXXX-X/18/06 CCS: Information systems Information retrieval Retrieval models and ranking Learning

Chaosheng Dong, Peiyao Xiao, Yijia Wang, Kaiyi Ji · 2026

Read DeepMTL2R: A Library for Deep Multi-task Learning to Rank Conference: Make sure to enter the correct conference title from your rights confirmation emai; June 03–05, 2018; Woodstock, NY ISBN: 978-1-4503-XXXX-X/18/06 CCS: Information systems Information retrieval Retrieval models and ranking Learning by Chaosheng Dong, Peiyao Xiao, Yijia Wang, Kaiyi Ji (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-14519.txt.

Description

Chaosheng Dong and three co-authors offer a 2026 arXiv open-source setup for deep multi-task learning to rank. DeepMTL2R tunes many fit rules at once, joining mixed fit signals into one joined, context-aware model via transformer self-focus. That design handles varied and at times clashing goals in new ranking systems.

The setup holds 21 top multi-task learning methods and backs multi-goal tuning to find Pareto-best ranking models. By grasping tangled ties and long-range links among items and labels, it offers a scalable, rich answer and enables fair weighs across multi-task plans. Tests on a public set show matching scores, with trade-offs among goals drawn out, and the code is public.

Catalog record

CREATORChaosheng Dong, Peiyao Xiao, Yijia Wang, Kaiyi Ji
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE15 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602-14519

Browse the full library · About the Tower of Babel