no code implementations • 16 May 2025 • Radek Osmulsk, Gabriel de Souza P. Moreira, Ronay Ak, Mengyao Xu, Benedikt Schifferer, Even Oldridge
MIRACL-VISION is a challenging, representative, multilingual evaluation benchmark for visual retrieval pipelines and will help the community build robust models for document retrieval.
no code implementations • 12 Sep 2024 • Gabriel de Souza P. Moreira, Ronay Ak, Benedikt Schifferer, Mengyao Xu, Radek Osmulski, Even Oldridge
We focus on text retrieval for question-answering tasks, a common use case for Retrieval-Augmented Generation systems.
no code implementations • 22 Jul 2024 • Gabriel de Souza P. Moreira, Radek Osmulski, Mengyao Xu, Ronay Ak, Benedikt Schifferer, Even Oldridge
One of the challenging aspects of fine-tuning embedding models is the selection of high quality hard-negative passages for contrastive learning.
1 code implementation • 20 Apr 2023 • Patrick John Chia, Giuseppe Attanasio, Jacopo Tagliabue, Federico Bianchi, Ciro Greco, Gabriel de Souza P. Moreira, Davide Eynard, Fahd Husain
Recommender Systems today are still mostly evaluated in terms of accuracy, with other aspects beyond the immediate relevance of recommendations, such as diversity, long-term user retention and fairness, often taking a back seat.
1 code implementation • 12 Jul 2022 • Jacopo Tagliabue, Federico Bianchi, Tobias Schnabel, Giuseppe Attanasio, Ciro Greco, Gabriel de Souza P. Moreira, Patrick John Chia
Much of the complexity of Recommender Systems (RSs) comes from the fact that they are used as part of more complex applications and affect user experience through a varied range of user interfaces.
no code implementations • 11 Jul 2021 • Gabriel de Souza P. Moreira, Sara Rabhi, Ronay Ak, Md Yasin Kabir, Even Oldridge
Session-based recommendation is an important task for e-commerce services, where a large number of users browse anonymously or may have very distinct interests for different sessions.
no code implementations • 22 Jun 2020 • Gabriel de Souza P. Moreira, Dietmar Jannach, Adilson Marques da Cunha
We describe a hybrid meta-architecture -- the CHAMELEON -- for session-based news recommendation that is able to leverage a variety of information types using Recurrent Neural Networks.
2 code implementations • 12 Jul 2019 • Gabriel de Souza P. Moreira, Dietmar Jannach, Adilson Marques da Cunha
A particular problem in that context is that online readers are often anonymous, which means that this personalization can only be based on the last few recorded interactions with the user, a setting named session-based recommendation.
3 code implementations • 31 Jul 2018 • Gabriel de Souza P. Moreira, Felipe Ferreira, Adilson Marques da Cunha
This architecture is composed of two modules, the first responsible to learn news articles representations, based on their text and metadata, and the second module aimed to provide session-based recommendations using Recurrent Neural Networks.