MEXMA: Token-level objectives improve sentence representations

João Maria Janeiro, Benjamin Piwowarski, Patrick Gallinari, Loïc Barrault
9/19/2024
cs.CLcs.AI

No official code found. Best implementation: facebookresearch/mexma (44 stars, MIT, last commit 2025-01-06).

Abstract

Current pre-trained cross-lingual sentence encoders approaches use sentence-level objectives only. This can lead to loss of information, especially for tokens, which then degrades the sentence representation. We propose MEXMA, a novel approach that integrates both sentence-level and token-level objectives. The sentence representation in one language is used to predict masked tokens in another language, with both the sentence representation and all tokens directly updating the encoder. We show that adding token-level objectives greatly improves the sentence representation quality across several tasks. Our approach outperforms current pre-trained cross-lingual sentence encoders on bi-text mining as well as several downstream tasks. We also analyse the information encoded in our tokens, and how the sentence representation is built from them.

View on arXivView PDF

Code Implementations(1)

facebookresearch/mexmaCommunity100%
MIT

MEXMA: Token-level objectives improve sentence representations

444PythonSep 11, 20241 years agoMIT

Similar Papers

Modular Sentence Encoders: Separating Language Specialization from Cross-Lingual Alignment

81% match1 repo82024

SentenceVAE: Enable Next-sentence Prediction for Large Language Models with Faster Speed, Higher Accuracy and Longer Context

80% match2024

Omnilingual SONAR: Cross-Lingual and Cross-Modal Sentence Embeddings Bridging Massively Multilingual Text and Speech

80% match2026

MAGNET: Improving the Multilingual Fairness of Language Models with Adaptive Gradient-Based Tokenization

80% match2026

SupraTok: Cross-Boundary Tokenization for Enhanced Language Model Performance

79% match2025

Token Prepending: A Training-Free Approach for Eliciting Better Sentence Embeddings from LLMs

79% match2024

Cite this paper

@article{janeiro2024mexma,
  title  = {MEXMA: Token-level objectives improve sentence representations},
  author = {João Maria Janeiro and Benjamin Piwowarski and Patrick Gallinari and Loïc Barrault},
  year   = {2024},
  eprint = {2409.12737},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2409.12737v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2409.12737.svg)](https://codeofpaper.com/paper/2409.12737)

Discussion