Goldilocks: Just-right tuning of bert for technology-assisted review

E Yang, S MacAvaney, DD Lewis, O Frieder - European Conference on …, 2022 - Springer
Technology-assisted review (TAR) refers to iterative active learning workflows for document
review in high recall retrieval (HRR) tasks. TAR research and most commercial TAR …

[HTML][HTML] Improved risk minimization algorithms for technology-assisted review

A Molinari, A Esuli, F Sebastiani - Intelligent Systems with Applications, 2023 - Elsevier
MINECORE is a recently proposed decision-theoretic algorithm for technology-assisted
review that attempts to minimise the expected costs of review for responsiveness and …

RLStop: A Reinforcement Learning Stopping Method for TAR

R Bin-Hezam, M Stevenson - arXiv preprint arXiv:2405.02525, 2024 - arxiv.org
We present RLStop, a novel Technology Assisted Review (TAR) stopping rule based on
reinforcement learning that helps minimise the number of documents that need to be …

Leveraging large-scale multimedia datasets to refine content moderation models

I Sarridis, C Koutlis, O Papadopoulou… - 2022 IEEE Eighth …, 2022 - ieeexplore.ieee.org
The sheer volume of online user-generated content has rendered content moderation
technologies essential in order to protect digital platform audiences from content that may …

[PDF][PDF] Intelligent Systems with Applications

MINECORE is a recently proposed decision-theoretic algorithm for technology-assisted
review that attempts to minimise the expected costs of review for responsiveness and …

Report on the 2nd international conference on design of experimental search & information retrieval systems (DESIRES 2021)

O Alonso, S Marchesin, M Najork, G Silvello - ACM SIGIR Forum, 2022 - dl.acm.org
This is a report on the second edition of the International Conference on Design of
Experimental Search & Information REtrieval Systems (DESIRES 2021) held at the …