Tomato, Tomahto, Tomate: Do Multilingual Language Models Understand Based on Subword-Level Semantic Concepts? (2025.findings-naacl)
Copied to clipboard
| Challenge: | a recent study shows that human understanding of text depends on general semantic concepts of words that are robust to their superficial forms. |
| Approach: | They evaluate the accuracy of multilingual multilingual language models based on subword-level semantics . they form "semantic tokens" by merging semantically similar subwords and embeddings based upon the results . |
| Outcome: | The proposed models are able to make predictions on multilingual tasks with different tokenizers and model sizes. |
Similar Papers
A Tour of Explicit Multilingual Semantics: Word Sense Disambiguation, Semantic Role Labeling and Semantic Parsing (2022.aacl-tutorials)
Copied to clipboard
| Challenge: | a recent advent of pretrained language models has sparked a revolution in NLP . but, there are still questions about whether current approaches capture explicit, symbolic meaning . this tutorial will review efforts to tackle three key open problems in lexical and sentence-level semantics . |
| Approach: | This tutorial reviews recent efforts to shed light on meaning in NLP . it will focus on three key open problems in lexical and sentence-level semantics . |
| Outcome: | This tutorial reviews recent efforts to shed light on meaning in NLP . it focuses on three key open problems in lexical and sentence-level semantics . |
A Systematic Analysis of Subwords and Cross-Lingual Transfer in Multilingual Translation (2024.findings-naacl)
Copied to clipboard
| Challenge: | Multilingual modelling can improve machine translation for low-resource languages, partly through shared subword representations. |
| Approach: | They propose to use subword regularisation to promote synergy and BPE to facilitate cross-lingual transfer. |
| Outcome: | The proposed methods promote synergy and prevent interference across different linguistic typologies. |
Tokenization Impacts Multilingual Language Modeling: Assessing Vocabulary Allocation and Overlap Across Languages (2023.findings-acl)
Copied to clipboard
| Challenge: | Multilingual language models perform surprisingly well in a variety of NLP tasks for diverse languages. |
| Approach: | They propose to evaluate the quality of lexical representation and vocabulary overlap observed in sub-word tokenizers. |
| Outcome: | The proposed criteria show that the overlap of vocabulary across languages can be detrimental to certain downstream tasks. |
Towards a Common Understanding of Contributing Factors for Cross-Lingual Transfer in Multilingual Language Models: A Review (2023.acl-long)
Copied to clipboard
| Challenge: | Pre-trained Multilingual Language Models have shown a strong ability to transfer knowledge across languages. |
| Approach: | They examine factors contributing to the ability of MLLMs to perform zero-shot cross-lingual transfer . they identify consensuses among studies with consistent findings and resolve conflicts . |
| Outcome: | The authors outline and discuss factors that contribute to the ability of MLLMs to perform zero-shot cross-lingual transfer. |
MultiLexBATS: Multilingual Dataset of Lexical Semantic Relations (2024.lrec-main)
Copied to clipboard
Dagmar Gromann, Hugo Goncalo Oliveira, Lucia Pitarch, Elena-Simona Apostol, Jordi Bernad, Eliot Bytyçi, Chiara Cantone, Sara Carvalho, Francesca Frontini, Radovan Garabik, Jorge Gracia, Letizia Granata, Fahad Khan, Timotej Knez, Penny Labropoulou, Chaya Liebeskind, Maria Pia Di Buono, Ana Ostroški Anić, Sigita Rackevičienė, Ricardo Rodrigues, Gilles Sérasset, Linas Selmistraitis, Mahammadou Sidibé, Purificação Silvano, Blerina Spahiu, Enriketa Sogutlu, Ranka Stanković, Ciprian-Octavian Truică, Giedre Valunaite Oleskeviciene, Slavko Zitnik, Katerina Zdravkova
| Challenge: | Prior work has focused on analysing lexical semantic relations in word embeddings or probing pretrained language models (PLMs) with some exceptions. |
| Approach: | They propose to use a multilingual parallel dataset of lexical semantic relations adapted from BATS in 15 languages including low-resource languages such as Bambara, Lithuanian, and Albanian as an experiment on cross-lingual transfer of relational knowledge. |
| Outcome: | The proposed dataset is adapted from a BATS-based dataset in 15 languages including low-resource languages such as Bambara, Lithuanian, and Albanian. |
MINERS: Multilingual Language Models as Semantic Retrievers (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing benchmarks have evaluated language models to evaluate their performance across a range of embedding tasks. |
| Approach: | They propose a benchmark to evaluate the robustness of multilingual language models in semantic retrieval tasks including bitext mining and classification via retrieval-augmented contexts. |
| Outcome: | The proposed framework evaluates the robustness of multilingual LMs in retrieval tasks across over 200 languages, including extremely low-resource languages in challenging cross-lingual and code-switching settings. |
Do Language Models Have Semantics? On the Five Standard Positions (2025.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) are trained to solve the so-called cloze task . solving clozing tasks is essentially a memorization task, says a recent study . |
| Approach: | They propose to use five positions to determine whether large language models exhibit semantic understanding . large language model is trained to solve the so-called cloze task . |
| Outcome: | The proposed theory is based on a pairwise comparison of five positions on semantic understanding in large language models and chatbots. |
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 . |
A Multi-dimensional Evaluation of Tokenizer-free Multilingual Pretrained Models (2023.findings-eacl)
Copied to clipboard
| Challenge: | Recent work on tokenizer-free models shows promising results in cross-lingual transfer . previous work focused on reporting accuracy on a limited set of tasks and data settings . |
| Approach: | They compare tokenizer-free and subword-based models using various dimensions . they find subword models are still the most practical choice in many settings . |
| Outcome: | The proposed model improves cross-lingual transfer and reduces engineering overhead. |
Analyzing the Evaluation of Cross-Lingual Knowledge Transfer in Multilingual Language Models (2024.eacl-long)
Copied to clipboard
| Challenge: | Recent advances in training multilingual models on large datasets have shown promising results in knowledge transfer across languages. |
| Approach: | They challenge the assumption that high zero-shot performance reflects high cross-lingual ability by introducing more challenging setups involving instances with multiple languages. |
| Outcome: | The proposed model can achieve high performance on multilingual benchmarks and on low-resource languages. |