Papers by Benjamin Muller
First Align, then Predict: Understanding the Cross-Lingual Ability of Multilingual BERT (2021.eacl-main)
Copied to clipboard
| Challenge: | Multilingual pretrained language models have demonstrated remarkable zero-shot cross-lingual transfer capabilities. |
| Approach: | They propose to use a layer ablation technique to create a multilingual model that is viewed as a stacking of two sub-networks: a language-agnostic encoder and a task-specific predictor. |
| Outcome: | The proposed model can perform zero-shot cross-lingual transfer for many languages. |
Establishing a New State-of-the-Art for French Named Entity Recognition (2020.lrec-1)
Copied to clipboard
| Challenge: | Named entity recognition (NER) is a task consisting in identifying text spans that denote named entities such as person, location and organization names. |
| Approach: | They manually annotated the French TreeBank with information related to named entities . they sketch the underlying annotation guidelines and provide a few figures about the annotations . |
| Outcome: | The French TreeBank is the main source of morphosyntactic and syntactical annotations for French. |
Building a User-Generated Content North-African Arabizi Treebank: Tackling Hell (2020.acl-main)
Copied to clipboard
Djamé Seddah, Farah Essaidi, Amal Fethi, Matthieu Futeral, Benjamin Muller, Pedro Javier Ortiz Suárez, Benoît Sagot, Abhishek Srivastava
| Challenge: | a treebank for a north-African Arabic dialect known for code-switching is made freely available . authors: geopolitical events are a factor highlighting a language deficiency in terms of natural language processing resources . |
| Approach: | They propose to make a treebank for a romanized user-generated content variety of Algerian . they supplement it with 50k unlabeled sentences from common crawl and web-crawled data . |
| Outcome: | The proposed treebank is made of 1500 sentences, fully annotated in morpho-syntax and universal dependency syntax, with full translation at both the word and sentence levels. |
Cross-Lingual Open-Domain Question Answering with Answer Sentence Generation (2022.aacl-main)
Copied to clipboard
| Challenge: | Open-Domain Generative Question Answering has achieved impressive performance in English . combining document-level retrieval with answer generation can generate complete sentences . |
| Approach: | They propose an open-domain approach that combines document retrieval with answer generation to generate complete sentences in English . they propose a cross-lingual generative model that exploits passages written in multiple languages . |
| Outcome: | The proposed model outperforms answer sentence selection baselines for all 5 languages and monolingual pipelines for three out of five languages. |
Evaluating and Modeling Attribution for Cross-Lingual Question Answering (2023.emnlp-main)
Copied to clipboard
Benjamin Muller, John Wieting, Jonathan Clark, Tom Kwiatkowski, Sebastian Ruder, Livio Soares, Roee Aharoni, Jonathan Herzig, Xinyi Wang
| Challenge: | Open-retrieval question answering systems are lacking in attribution for cross-lingual question answering . open-research questions are available in 20 languages, but their raw generation often falls short in factuality . |
| Approach: | They are the first to study attribution for cross-lingual question answering . they collect data in 5 languages to assess the attribution level of a state-of-the-art QA system . |
| Outcome: | The proposed approach improves the attribution level of a state-of-the-art cross-lingual QA system. |
SpiRit-LM: Interleaved Spoken and Written Language Model (2025.tacl-1)
Copied to clipboard
Tu Anh Nguyen, Benjamin Muller, Bokai Yu, Marta R. Costa-jussa, Maha Elbayad, Sravya Popuri, Christophe Ropers, Paul-Ambroise Duquenne, Robin Algayres, Ruslan Mavlyutov, Itai Gat, Mary Williamson, Gabriel Synnaeve, Juan Pino, Benoît Sagot, Emmanuel Dupoux
| Challenge: | SpiRit-LM is a foundation multimodal language model that freely mixes text and speech. |
| Approach: | They propose a multimodal language model that freely mixes text and speech . they extend the model to the speech modality by continuously training it on text and language units. |
| Outcome: | The proposed model can learn new tasks in a few-shot fashion across modalities. |
The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants (2024.acl-long)
Copied to clipboard
Lucas Bandarkar, Davis Liang, Benjamin Muller, Mikel Artetxe, Satya Narayan Shukla, Donald Husa, Naman Goyal, Abhinandan Krishnan, Luke Zettlemoyer, Madian Khabsa
| Challenge: | Existing benchmarks for text comprehension only cover 30 languages, but lack of labeled data is a major obstacle to building functional systems in most languages. |
| Approach: | They present a multiple-choice machine reading comprehension dataset spanning 122 languages . they use it to evaluate the capabilities of multilingual masked language models and large language models . |
| Outcome: | The proposed dataset enables the evaluation of text models in high-, medium- and low-resource languages. |
In What Languages are Generative Language Models the Most Formal? Analyzing Formality Distribution across Languages (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Multilingual generative language models (LMs) are fluent in a large variety of languages. |
| Approach: | They analyze formality distributions of XGLM and BLOOM’s predictions in 5 languages and classify 1,200 generations per language as formal, informal, or incohesive. |
| Outcome: | The proposed models generate a significant amount of informal predictions even when prompted with formal text. |
CamemBERT: a Tasty French Language Model (2020.acl-main)
Copied to clipboard
Louis Martin, Benjamin Muller, Pedro Javier Ortiz Suárez, Yoann Dupont, Laurent Romary, Éric de la Clergerie, Djamé Seddah, Benoît Sagot
| Challenge: | Pretrained language models are now ubiquitous in Natural Language Processing, but their use in other languages is limited. |
| Approach: | They propose to train monolingual Transformer-based model for other languages using web crawled data instead of Wikipedia data and a relatively small web crawl dataset leads to better results. |
| Outcome: | The proposed model performs as well as those obtained using larger datasets. |
Enhancing BERT for Lexical Normalization (D19-55)
Copied to clipboard
| Challenge: | Pre-trained contextual language models have improved performance of many NLP tasks. |
| Approach: | They propose to use a pre-trained language model to perform lexical normalisation without UGC resources. |
| Outcome: | The proposed model can perform lexical normalisation without the need for training sentences and 3,000 tokens. |
Byte Latent Transformer: Patches Scale Better Than Tokens (2025.acl-long)
Copied to clipboard
Artidoro Pagnoni, Ramakanth Pasunuru, Pedro Rodriguez, John Nguyen, Benjamin Muller, Margaret Li, Chunting Zhou, Lili Yu, Jason E Weston, Luke Zettlemoyer, Gargi Ghosh, Mike Lewis, Ari Holtzman, Srini Iyer
| Challenge: | Existing large language models (LLMs) are trained on bytes, except for tokenization, which groups bytes into a static set of tokens. |
| Approach: | They propose a new byte-level LLM architecture that encodes bytes into dynamically sized patches, which serve as the primary units of computation. |
| Outcome: | The proposed architecture matches tokenization-based models with improvements in inference efficiency and robustness. |
When Being Unseen from mBERT is just the Beginning: Handling New Languages With Multilingual Language Models (2021.naacl-main)
Copied to clipboard
| Challenge: | Language models are a new standard to build state-of-the-art NLP systems. |
| Approach: | They compare multilingual and monolingual models on unseen languages . they show that some languages benefit from transfer learning whereas others don't . |
| Outcome: | The proposed model behaves in multiple ways on unseen languages, while others fail to transfer . the results provide a promising direction towards making multilingual models useful for a new set of unseense languages. |