OWL: Probing Cross-Lingual Recall of Memorized Texts via World Literature (2025.emnlp-main)
Copied to clipboard
Alisha Srivastava, Emir Kaan Korukluoglu, Minh Nhat Le, Duyen Tran, Chau Minh Pham, Marzena Karpinska, Mohit Iyyer
| Challenge: | Large language models (LLMs) are known to memorize and recall English text from their pretraining data, but the extent to which this ability generalizes to non-English languages or transfers across languages remains unclear. |
| Approach: | They propose a dataset of 31.5K aligned excerpts from 20 books in ten languages, including English originals, official translations and new translations in six low-resource languages. |
| Outcome: | The proposed model can recall English content in translations, but perturbations reduce performance, causing the model to fail. |
Similar Papers
Shared Path: Unraveling Memorization in Multilingual LLMs through Language Similarities (2025.emnlp-main)
Copied to clipboard
| Challenge: | Using multilingual models, we find that treating languages in isolation obscures the true patterns of memorization. |
| Approach: | They propose a graph-based correlation metric that incorporates language similarity to analyze cross-lingual memorization. |
| Outcome: | The proposed model incorporates language similarity to analyze cross-lingual memorization in 95 languages. |
How Do Multilingual Language Models Remember Facts? (2025.findings-acl)
Copied to clipboard
| Challenge: | Prior research has focused on English monolingual models, but how these mechanisms generalize to non-English languages remains unexplored. |
| Approach: | They analyze three multilingual LLMs to find out how they can generalize recall mechanisms . they find that subject enrichment is language-independent, object extraction is language dependent . |
| Outcome: | The proposed model performs better in multilingual contexts than in English models . the model is more efficient in multi-lingual context, but it is more complex in multilinguistic models compared to English models. |
Probing the Emergence of Cross-lingual Alignment during LLM Training (2024.findings-acl)
Copied to clipboard
| Challenge: | Multilingual Large Language Models (LLMs) achieve remarkable levels of zero-shot cross-lingual transfer performance. |
| Approach: | They propose that LLMs can align languages without explicit supervision from parallel sentences without a single linguistic feature. |
| Outcome: | The proposed model can perform zero-shot cross-lingual transfer even when the vocabularies of two languages have a null intersection, i.e., no tokens are shared. |
Multilingual Amnesia: On the Transferability of Unlearning in Multilingual LLMs (2026.eacl-long)
Copied to clipboard
Alireza Dehghanpour Farashah, Aditi Khandelwal, Marylou Fauchard, Zhuan Shi, Negar Rostamzadeh, Golnoosh Farnadi
| Challenge: | Existing studies on unlearning in multilingual large language models focus on monolingual settings, typically English. |
| Approach: | They propose to use a multilingual data and concept unlearning model to investigate the problem . they extend benchmarks for factual knowledge and stereotypes into ten languages . |
| Outcome: | The proposed model is able to unlearning in 10 languages across five languages and resource levels. |
Don’t Trust ChatGPT when your Question is not in English: A Study of Multilingual Abilities and Types of LLMs (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies have shown that large language models can perform a wide variety of language tasks when presented in English. |
| Approach: | They propose a method to evaluate the multilingual capabilities of large language models using a prompt back-translation method to find out how LLMs acquire their multilingual abilities. |
| Outcome: | The proposed method shows that large language models can transfer learned knowledge across different languages, but struggle to provide accurate results in translation-variant tasks. |
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. |
X-FACTR: Multilingual Factual Knowledge Retrieval from Pretrained Language Models (2020.emnlp-main)
Copied to clipboard
| Challenge: | Language models (LMs) capture factual knowledge by filling in the blanks of cloze-style prompts. |
| Approach: | They propose a code-switching-based method to improve the ability of multilingual LMs to access knowledge and verify its effectiveness on several benchmark languages. |
| Outcome: | The proposed method improves the ability of multilingual LMs to access knowledge and verify its effectiveness on several benchmark languages. |
Tracing Multilingual Factual Knowledge Acquisition in Pretraining (2025.findings-emnlp)
Copied to clipboard
Yihong Liu, Mingyang Wang, Amir Hossein Kargaran, Felicia Körner, Ercong Nie, Barbara Plank, François Yvon, Hinrich Schuetze
| Challenge: | Large Language Models are capable of recalling multilingual factual knowledge, but most studies evaluate only the final model, leaving the development of factual recall and crosslingual consistency unexplored. |
| Approach: | They trace how factual recall and crosslingual consistency evolve during pretraining, focusing on OLMo-7B as a case study. |
| Outcome: | The results show that fact frequency is the key to a better recall of multilingual facts, regardless of language, and some low-frequency facts in non-English languages can still be correctly recalled. |
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. |
Language Directions in Multilingual LLMs: A Layer-wise Diagnostic Study of Token Alignment and Pretraining Imprint (2026.acl-srw)
Copied to clipboard
| Challenge: | Using a unified probing framework, we analyze six multilingual LLMs across five languages. |
| Approach: | They analyze multilingual representations across five languages and analyze their behavior . they find that accuracy rises by +73.5 to +80.7 points from L0 to L1 on average . |
| Outcome: | The proposed framework enables a consistent and substantial early jump in accuracy across models . the token–language alignment measures where vocabulary sharing peaks . |