Home Research Papers Datasets State of the Art Pricing

Discover, visualize, and connect AI research papers. Explore the latest trends and insights in artificial intelligence research.

Product

Home
Research Papers
About

Support

Contact
Terms of Service
Privacy Policy

© 2026 Papersgraph. All rights reserved.

natural-language-processing-2

Cross-Lingual Natural Language Inference

3260 papers • 126 benchmarks • 313 datasets

Using data and models available for one language for which ample such resources are available (e.g., English) to solve a natural language inference task in another, commonly more low-resource, language.

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in cross-lingual-natural-language-inference-4

Trend

Dataset

Best Model

Actions

XNLI

XNLI

XNLI Zero-Shot English-to-Spanish

XNLI Zero-Shot English-to-Spanish

Libraries

i

Use these libraries to find cross-lingual-natural-language-inference-4 models and implementations

2 papers 391

Datasets

XNLI

XGLUE

Subtasks

No subtasks available.

Most implemented papers

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Jacob Devlin, Kenton Lee, Kristina Toutanova, Ming-Wei Chang•Mon Dec 31 2018

A new language representation model, BERT, designed to pre-train deep bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers, which can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of tasks.

109344

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

XNLI Zero-Shot English-to-German

XNLI Zero-Shot English-to-German

XNLI Zero-Shot English-to-French

XNLI Zero-Shot English-to-French

0

Supervised Learning of Universal Sentence Representations from Natural Language Inference Data

Antoine Bordes, Alexis Conneau, Douwe Kiela, Holger Schwenk, Loïc Barrault•Thu May 04 2017

It is shown how universal sentence representations trained using the supervised data of the Stanford Natural Language Inference datasets can consistently outperform unsupervised methods like SkipThought vectors on a wide range of transfer tasks.

2173 0

ByT5: Towards a Token-Free Future with Pre-trained Byte-to-Byte Models

Sharan Narang, Rami Al-Rfou, Adam Roberts, Colin Raffel, Noah Constant, Linting Xue, Mihir Kale, Aditya Barua•Thu May 27 2021

This paper shows that a standard Transformer architecture can be used with minimal modifications to process byte sequences, characterize the trade-offs in terms of parameter count, training FLOPs, and inference speed, and shows that byte-level models are competitive with their token-level counterparts.

614 0

Better Fine-Tuning by Reducing Representational Collapse

Luke Zettlemoyer, S. Gupta, Naman Goyal, Anchit Gupta, Armen Aghajanyan, Akshat Shrivastava•Wed Aug 05 2020

A simplified and efficient method rooted in trust region theory that replaces previously used adversarial objectives with parametric noise (sampling from either a normal or uniform distribution), thereby discouraging representation change during fine-tuning when possible without hurting performance.

228 0

XNLI: Evaluating Cross-lingual Sentence Representations

Alexis Conneau, Holger Schwenk, Samuel R. Bowman, Adina Williams, Guillaume Lample, Ruty Rinott, Veselin Stoyanov•Wed Sep 12 2018

This work constructs an evaluation set for XLU by extending the development and test sets of the Multi-Genre Natural Language Inference Corpus to 14 languages, including low-resource languages such as Swahili and Urdu and finds that XNLI represents a practical and challenging evaluation suite and that directly translating the test data yields the best performance among available baselines.

1536 0

Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond

Holger Schwenk, Mikel Artetxe•Tue Dec 25 2018

An architecture to learn joint multilingual sentence representations for 93 languages, belonging to more than 30 different families and written in 28 different scripts using a single BiLSTM encoder with a shared byte-pair encoding vocabulary for all languages, coupled with an auxiliary decoder and trained on publicly available parallel corpora.

1102 0

Rethinking embedding coupling in pre-trained language models

Melvin Johnson, Sebastian Ruder, Hyung Won Chung, Thibault Févry, Henry Tsai•Fri Oct 23 2020

The analysis shows that larger output embeddings prevent the model's last layers from overspecializing to the pre-training task and encourage Transformer representations to be more general and more transferable to other tasks and languages.

170 0

Language Embeddings for Typology and Cross-lingual Transfer Learning

Dian Yu, Taiqi He, Kenji Sagae•Wed Jun 02 2021

This work generates dense embeddings for 29 languages using a denoising autoencoder, and evaluates the embedDings using the World Atlas of Language Structures (WALS) and two extrinsic tasks in a zero-shot setting: cross-lingual dependency parsing and cross-lingsual natural language inference.

13 0

PARADISE: Exploiting Parallel Data for Multilingual Sequence-to-Sequence Pretraining

Mikel Artetxe, Machel Reid•Tue Aug 03 2021

PARADISE (PARAllel & Denoising Integration in SEquence-to-sequence models), which extends the conventional denoising objective used to train these models by replacing words in the noised sequence according to a multilingual dictionary, and predicting the reference translationaccording to a parallel corpus instead of recovering the original sequence.

30 0

Subword Mapping and Anchoring across Languages

Giorgos Vernikos, Andrei Popescu-Belis•Wed Sep 08 2021

The benefits of SMALA for cross-lingual natural language inference (XNLI), where it improves zero-shot transfer to an unseen language without task-specific data, but only by sharing subword embeddings, and in neural machine translation, it shows that joint subword vocabularies obtained with SMAL a lead to higher BLEU scores on sentences that contain many false positives and false negatives.

13 0

Adding a benchmark result helps the community track progress.

Cross-Lingual Natural Language Inference | State-of-the-Art