Multilingual Virtual Agents for Customer Service MAIA

Project: Multilingual Virtual Agents for Customer Service

Acronym: MAIA

File: cartaz_vertical_maia.pdf

Main Objective:
Create multilingual virtual agent assistants for customer service by combining new natural language processing and machine learning techniques for machine translation and dialogue applications.

Region of Intervention: Lisbon

Reference: 045909

Funding: P2020

Approval Date: 01-07-2020

Start Date: 01-07-2020

End Date: 01-07-2023

Team: André Filipe Torres Martins, Taisiya Glushkova, Mario Alexandre Teles de Figueiredo, Nuno Miguel Serrano Guerreiro, Patrick Santos Fernandes, António José Evaristo Farinhas, .

Groups: Pattern and Image Analysis – Lx

Partners: Carnegie Mellon University, INESC-ID, Unbabel

Local Coordinator: André Filipe Torres Martins

United Nations Strategic Development Goal - Industry, Innovation and Infrastructure

Associated Publications

3Papers in Conferences
T. Glushkova, C Zerva, A. Martins, BLEU Meets COMET: Combining Lexical and Neural Metrics Towards Robust Machine Translation Evaluation, Conference of the European Association for Machine Translation, Tampere, Finland, June, 2023,
| Abstract | Full text (PDF 305 KBs) | BibTex
N. Guerreiro, E. Voita, A. Martins, Looking for a Needle in a Haystack: A Comprehensive Study of Hallucinations in Neural Machine Translation, EACL: Conference of the European Chapter of the Association for Computational Linguistics EACL, Dubrovnik, Croatia, Vol., pp. -, May, 2023,
| Abstract | Full text (PDF 1 MB) | BibTex
P. Fernandes, M. V. T. Treviso, D. Pruthi, A. Martins, G. Neubig, Learning to Scaffold: Optimizing Model Explanations for Teaching, Advances in Neural Information Processing Systems - NIPS, New Orleans, United States, Vol., pp. -, November, 2022,
| Abstract | BibTex

Project / MAIA

Project: Multilingual Virtual Agents for Customer Service