no code implementations • 31 Mar 2024 • Nathan Cornille, Marie-Francine Moens, Florian Mai
By training to predict the next token in an unlabeled corpus, large language models learn to perform many tasks without any labeled data.
1 code implementation • 19 Feb 2024 • Justus-Jonas Erker, Florian Mai, Nils Reimers, Gerasimos Spanakis, Iryna Gurevych
Search-based dialog models typically re-encode the dialog history at every turn, incurring high cost.
2 code implementations • 29 May 2023 • Florian Mai, Juan Zuluaga-Gomez, Titouan Parcollet, Petr Motlicek
In particular, multi-head HyperConformer achieves comparable or higher recognition performance while being more efficient than Conformer in terms of inference speed, memory, parameter count, and available training data.
2 code implementations • NeurIPS 2023 • Darko Drakulic, Sofia Michel, Florian Mai, Arnaud Sors, Jean-Marc Andreoli
In this paper, we present a novel formulation of Combinatorial Optimization Problems (COPs) as Markov Decision Processes (MDPs) that effectively leverages common symmetries of COPs to improve out-of-distribution robustness.
Combinatorial Optimization Out-of-Distribution Generalization
3 code implementations • 7 Mar 2022 • Florian Mai, Arnaud Pannatier, Fabio Fehr, Haolin Chen, Francois Marelli, Francois Fleuret, James Henderson
We find that existing architectures such as MLPMixer, which achieves token mixing through a static MLP applied to each feature independently, are too detached from the inductive biases required for natural language understanding.
no code implementations • 13 Oct 2021 • Florian Mai, James Henderson
We address this issue by extending their method to Bag-of-Vectors Autoencoders (BoV-AEs), which encode the text into a variable-size bag of vectors that grows with the size of the text, as in attention-based models.
1 code implementation • EMNLP 2020 • Florian Mai, Nikolaos Pappas, Ivan Montero, Noah A. Smith, James Henderson
Text autoencoders are commonly used for conditional generation tasks such as style transfer.
no code implementations • ICML 2020 • Prabhu Teja Sivaprasad, Florian Mai, Thijs Vogels, Martin Jaggi, François Fleuret
The performance of optimizers, particularly in deep learning, depends considerably on their chosen hyperparameter configuration.
1 code implementation • 22 Jul 2019 • Lukas Galke, Florian Mai, Iacopo Vagliano, Ansgar Scherp
We present multi-modal adversarial autoencoders for recommendation and evaluate them on two different tasks: citation recommendation and subject label recommendation.
1 code implementation • ICLR 2019 • Florian Mai, Lukas Galke, Ansgar Scherp
In order to address this shortcoming, we propose a learning algorithm for the Continuous Matrix Space Model, which we call Continual Multiplication of Words (CMOW).
1 code implementation • 15 May 2017 • Lukas Galke, Florian Mai, Alan Schelten, Dennis Brunsch, Ansgar Scherp
For the first time, we offer a systematic comparison of classification approaches to investigate how far semantic annotations can be conducted using just the metadata of the documents such as titles published as labels on the Linked Open Data cloud.