Multilingual Universal Sentence Encoder for Semantic Retrieval

By Yinfei Yang, Daniel Matthew Cer, Amin Ahmad, Mandy Guo, Jax Law, Noah Constant, Gustavo Hernández Ábrego, Steve Yuan, Chris Tar, Yun-Hsuan Sung and 2 more. 2019

534 citations recorded by the source corpus.

We present easy-to-use retrieval focused multilingual sentence embedding models, made available on TensorFlow Hub. The models embed text from 16 languages into a shared semantic space using a multi-task trained dual-encoder that learns tied cross-lingual representations via translation bridge tasks (Chidambaram et al., 2018). The models achieve a new state-of-the-art in performance on monolingual and cross-lingual semantic retrieval (SR). Competitive performance is obtained on the related tasks of translation pair bitext retrieval (BR) and retrieval question answering (ReQA). On transfer learning tasks, our multilingual embeddings approach, and in some cases exceed, the performance of English only sentence embeddings.

Subfields: Artificial Intelligence, Information Retrieval & Search, Natural Language Processing, Multimodal Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Information Retrieval & Search

In Natural Language Processing

In Multimodal Learning

Citations in the corpus

MURAL: Multimodal, Multitask Retrieval Across Languages (2021, methodology): “egy reported by Aggarwal and Kale (2020)— mUSE (Yang et al., 2020) with a multimodal metric loss (M3L).”
Crisscrossed Captions: Extended Intramodal and Intermodal Semantic Similarity Judgments for MS-COCO (2020, methodology): “We use the multilingual transformer version from TensorFlow Hub (Yang et al., 2020).”
XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning (2020, background): “Datasets that cover multiple languages for other natural understanding tasks, such as language inference (Conneau et al., 2018), question answering (Lewis et al., 2019; Artetxe et al., 2020a; Clark et al., 2020), and paraphrase identiﬁcation (Yang et al., 2019b) have received increasing attention.”
BC-Z: Zero-Shot Task Generalization with Robotic Imitation Learning (2022, methodology): “If the command is a language command, we use a pretrained multilingual sentence encoder [46]1 as our encoder, producing a 512-dim language vector for each task.”
Multilingual and Cross-Lingual Intent Detection from Spoken Data (2021, background): “mUSE (Yang et al., 2020) is a multilingual version of the Universal Sentence Encoder (USE) model for English (Cer et al.”
Cross-lingual and Multilingual CLIP (2022, background): “Finally, there has been work that fine-tunes multi-lingual sentence embedding models, such as mUSE (Yang et al., 2020) and LASER (Artetxe and Schwenk, 2019) to match a pre-trained vision embedding space (Aggarwal et al., 2021).”
MultiCQA: Zero-Shot Transfer of Self-Supervised Text Matching Models on a Massive Scale (2020, background): “They propose USE-QA, a sentence encoder for comparing questions and answers, and achieve promising zero-shot results in retrieval tasks.”
A Bilingual Generative Transformer for Semantic Sentence Embedding (2019, background): “pel, 2018) or bilingual data (Wieting et al., 2019b; Chidambaram et al., 2019; Yang et al., 2020).”
Large Dual Encoders Are Generalizable Retrievers (2021, background): “Typical neural retrieval models follow a dual encoder paradigm (Gillick et al., 2018; Yang et al., 2020; Karpukhin et al., 2020).”
Unsupervised Multilingual Sentence Embeddings for Parallel Corpus Mining (2020, background): “The universal sentence encoder (USE) (Cer et al., 2018; Yang et al., 2019) family covers sentence embedding models with a multi-task dual-encoder training framework including the tasks of question-answer prediction or natural language inference.”