Papers by José G. C. de Souza

2 papers
Quality Estimation for Automatically Generated Titles of eCommerce Browse Pages (N18-3)

Copied to clipboard

Challenge: We are generating millions of titles using machine translation, but they are prone to errors.
Approach: They propose a Random Forest model which explores hand-crafted features and new features . they also propose SNs which embed metadata and generated title in the same space .
Outcome: The proposed models outperform the existing models on in-house data.
Quality-Aware Decoding for Neural Machine Translation (2022.naacl-main)

Copied to clipboard

Challenge: Despite advances in machine translation quality estimation and evaluation, decoding is mostly oblivious to this.
Approach: They propose to use a decoding framework that is quality-aware for neural machine translation . they compare various methods like N-best reranking and minimum Bayes risk decoding .
Outcome: The proposed quality-aware decoding outperforms MAP-based decoding on four datasets and two model classes.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations