Challenge: POStag and dependency n-grams are more effective than text length and global linguistic indices for this kind of task.
Approach: They propose to use POStag and dependency n-grams to predict the quality of a text written by learners of another language to categorize texts according to their CEFR level.
Outcome: The proposed model is more effective than POStag and dependency n-grams in cross-lingual experiments than the previous models.

Similar Papers

UniversalCEFR: Enabling Open Multilingual Research on Language Proficiency Assessment (2025.emnlp-main)

Copied to clipboard

Challenge: Language proficiency research plays a central role in education and often intersects with advances in linguistics and AI.
Approach: They propose a multilingual multidimensional dataset of texts annotated according to the CEFR scale in 13 languages.
Outcome: The proposed dataset supports linguistic features and pretrained models in multilingual CEFR level assessment.
Understanding Cross-Lingual Alignment—A Survey (2024.findings-acl)

Copied to clipboard

Challenge: Cross-lingual alignment is the meaningful similarity of representations across languages in multilingual language models.
Approach: They propose a taxonomy of methods to improve cross-lingual alignment . they argue that an effective trade-off between language-neutral and language-specific information is key .
Outcome: The proposed methods can be applied to encoder models and encoder-decoder-only models . they show that language-neutral and language-specific information is key .
Using Multilingual Resources to Evaluate CEFRLex for Learner Applications (2020.lrec-1)

Copied to clipboard

Challenge: The Common European Framework of Reference for Languages defines six levels of learner proficiency and links them to particular communicative abilities.
Approach: They propose to compile lexical resources that link single words and multi-word expressions to specific CEFR levels.
Outcome: The results show that the English CEFRLex resource is in accordance with external resources that are gold standard.
Thesis Proposal: Targeted and Unified Cross-Lingual Unlearning from Multilingual Language Models (2026.acl-srw)

Copied to clipboard

Challenge: Large language models trained on corpora scraped from the web can reproduce sensitive and copyright-protected data.
Approach: They propose to extend existing benchmarks to multilingual data by compiling parallel translations of question-answer pairs consisting of real-world facts and synthetic personally identifiable information.
Outcome: The proposed dataset will include translations of question-answer pairs consisting of real-world facts and synthetic personally identifiable information.
A Robust Self-Learning Method for Fully Unsupervised Cross-Lingual Mappings of Word Embeddings: Making the Method Robustly Reproducible as Well (2020.lrec-1)

Copied to clipboard

Challenge: Existing methods for fully unsupervised cross-lingual mapping of word embeddings are available to achieve such a mapping .
Approach: They reproduce the experiments of Artetxe and Sgaard (2018) . they propose a robust self-learning method for fully unsupervised cross-lingual mappings of word embeddings.
Outcome: The proposed method is feasible with minor assumptions, and it is able to be replicated in four languages.
Cross-lingual Editing in Multilingual Language Models (2024.findings-eacl)

Copied to clipboard

Challenge: Existing models editing techniques (METs) can efficiently update outdated LLMs without retraining.
Approach: They propose a cross-lingual model editing paradigm where a fact is edited in one language and the subsequent update propagation is observed across other languages.
Outcome: The proposed techniques perform well in multilingual models with knowledge stored in multiple languages.
An SLA Corpus Annotated with Pedagogically Relevant Grammatical Structures (L18-1)

Copied to clipboard

Challenge: a study using a framework to evaluate a language learner's proficiency in a second language aims to examine the production of learners with pedagogically relevant grammatical structures .
Approach: They annotated texts produced by language learners with grammatical structures . they found that learners from different proficiency levels use pedagogically relevant structures compared to those of already certified language learners .
Outcome: The annotated resource SGATe analyzes texts produced by language learners with grammatical structures . structure evolution along levels and level in which they are used the most was studied .
Baselines and Test Data for Cross-Lingual Inference (L18-1)

Copied to clipboard

Challenge: Recent research on textual entailment is limited to English, but it is expanding to other languages.
Approach: They propose to extend the research in SNLI-style natural language inference toward multilingual evaluation by using cross-lingual word embeddings and machine translation.
Outcome: The proposed system scores an average accuracy of just over 75%, but it is not perfect.
A Call for More Rigor in Unsupervised Cross-lingual Learning (2020.acl-main)

Copied to clipboard

Challenge: Existing research on unsupervised cross-lingual learning has focused on purely unsupervised learning without any parallel data for most of the world's languages.
Approach: They propose to define "multilingual learning" as learning a common model for two or more languages from raw text, without any downstream task labels.
Outcome: The proposed model is based on a model with no parallel data and abundant monolingual data.
Cross-lingual Similarity of Multilingual Representations Revisited (2022.aacl-main)

Copied to clipboard

Challenge: Similarity indexes like CKA and CCA are not suitable for cross-lingual learning analysis.
Approach: They propose an alternative that is exempt from the difficulties of CKA/CCA and is good specifically in a cross-lingual context.
Outcome: The proposed method is exempt from the difficulties of CKA/CCA and is good specifically in a cross-lingual context.

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