ConceptRM : The Quest to Mitigate Alert Fatigue through Consensus-Based Purity-Driven Data Cleaning for Reflection Modelling

Yongda Yu, Lei Zhang, Xinxin Guo, Minghui Yu, Zhengqi Zhuang, Guoping Rong, Haifeng Shen, Zhengfeng Li, Boge Wang, Guoan Zhang, Bangyu Xiang, Xiaobin Xu · 2026

Read ConceptRM : The Quest to Mitigate Alert Fatigue through Consensus-Based Purity-Driven Data Cleaning for Reflection Modelling by Yongda Yu, Lei Zhang, Xinxin Guo, Minghui Yu, Zhengqi Zhuang, Guoping Rong, Haifeng Shen, Zhengfeng Li, Boge Wang, Guoan Zhang, Bangyu Xiang, Xiaobin Xu (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-2602-20166.txt.

Description

Agents flood users with alerts, mostly false. Users grow numb and miss the critical ones. That numbness is called alert fatigue. A 2026 arXiv paper by Yongda Yu and eleven co-authors proposes ConceptRM, a way to build clean training data for a reflection model that catches false alerts.

Reflection models normally learn as filters from user verification feedback. That feedback data is noisy because it comes from production use, and cleaning it by hand costs too much. ConceptRM starts from a small set of expert labels used as anchors. It builds noisy variants of the data at several noise levels. Several distinct models then learn together through co-teaching. Their agreement reveals which examples are clean. Those examples form the corpus for training the reflection model.

Catalog record

CREATORYongda Yu, Lei Zhang, Xinxin Guo, Minghui Yu, Zhengqi Zhuang, Guoping Rong, Haifeng Shen, Zhengfeng Li, Boge Wang, Guoan Zhang, Bangyu Xiang, Xiaobin Xu
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE44 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2602-20166

Browse the full library · About the Tower of Babel