no code implementations • 24 May 2023 • Najah Ghalyan, Kostis Gourgoulias, Yash Satsangi, Sean Moran, Maxime Labonne, Joseph Sabelja
This paper proposes a method to estimate the class separability of an unlabeled text dataset by inspecting the topological characteristics of sentence-transformer embeddings of the text.
1 code implementation • 3 Apr 2023 • Maxime Labonne, Sean Moran
Our results demonstrate that Spam-T5 surpasses baseline models and other LLMs in the majority of scenarios, particularly when there are a limited number of training samples available.
no code implementations • 4 Dec 2021 • Maxime Labonne, Charalampos Chatzinakis, Alexis Olivereau
Predicting the bandwidth utilization on network links can be extremely useful for detecting congestion in order to correct them before they occur.
no code implementations • 25 Aug 2021 • Jorge López, Maxime Labonne, Claude Poletti
To guarantee a proper data collection, verifying that the collected data set holds certain properties is a possible solution.
no code implementations • 29 Nov 2020 • Maxime Labonne, Jorge López, Claude Poletti, Jean-Baptiste Munier
Predictions can be fed to the management system instead of current flows bandwidth in order to take decisions on a more accurate network state.