1 code implementation • 25 Jan 2024 • Alireza Mohammadshahi, Ali Shaikh, Majid Yazdani
In this paper, we propose an architecture to harness the collective knowledge of multiple trained LLMs to create a new state-of-the-art.
Ranked #4 on Multi-task Language Understanding on MMLU (using extra training data)
1 code implementation • 13 Nov 2023 • Alireza Mohammadshahi, Jannis Vamvas, Rico Sennrich
Massively multilingual machine translation models allow for the translation of a large number of languages with a single model, but have limited performance on low- and very-low-resource translation directions.
1 code implementation • 27 Oct 2023 • James Henderson, Alireza Mohammadshahi, Andrei C. Coman, Lesly Miculicich
We argue that Transformers are essentially graph-to-graph models, with sequences just being a special case.
1 code implementation • 13 Sep 2023 • Rico Sennrich, Jannis Vamvas, Alireza Mohammadshahi
Experiments on the massively multilingual models M2M-100 (418M) and SMaLL-100 show that these methods suppress hallucinations and off-target translations, reducing the number of translations with segment-level chrF2 below 10 by 67-83% on average, and the number of translations with oscillatory hallucinations by 75-92% on average, across 57 tested translation directions.
1 code implementation • 2 Nov 2022 • Alireza Mohammadshahi, Thomas Scialom, Majid Yazdani, Pouya Yanki, Angela Fan, James Henderson, Marzieh Saeidi
We demonstrate that RQUGE has a higher correlation with human judgment without relying on the reference question.
3 code implementations • 20 Oct 2022 • Alireza Mohammadshahi, Vassilina Nikoulina, Alexandre Berard, Caroline Brun, James Henderson, Laurent Besacier
In recent years, multilingual machine translation models have achieved promising performance on low-resource language pairs by sharing information between similar languages, thus enabling zero-shot translation.
no code implementations • *SEM (NAACL) 2022 • Luis Espinosa-Anke, Alexander Shvets, Alireza Mohammadshahi, James Henderson, Leo Wanner
Recognizing and categorizing lexical collocations in context is useful for language learning, dictionary compilation and downstream NLP.
1 code implementation • 22 May 2022 • Alireza Mohammadshahi, Vassilina Nikoulina, Alexandre Berard, Caroline Brun, James Henderson, Laurent Besacier
In this work, we assess the impact of compression methods on Multilingual Neural Machine Translation models (MNMT) for various language groups, gender, and semantic biases by extensive analysis of compressed models on different machine translation benchmarks, i. e. FLORES-101, MT-Gender, and DiBiMT.
1 code implementation • ACL (IWPT) 2021 • James Barry, Alireza Mohammadshahi, Joachim Wagner, Jennifer Foster, James Henderson
The task involves parsing Enhanced UD graphs, which are an extension of the basic dependency trees designed to be more facilitative towards representing semantic structure.
no code implementations • 15 Apr 2021 • Alireza Mohammadshahi, James Henderson
Recent models have shown that incorporating syntactic knowledge into the semantic role labelling (SRL) task leads to a significant improvement.
Ranked #6 on Semantic Role Labeling on CoNLL 2005 (using extra training data)
1 code implementation • 29 Mar 2020 • Alireza Mohammadshahi, James Henderson
We propose the Recursive Non-autoregressive Graph-to-Graph Transformer architecture (RNGTr) for the iterative refinement of arbitrary graphs through the recursive application of a non-autoregressive Graph-to-Graph Transformer and apply it to syntactic dependency parsing.
Ranked #8 on Dependency Parsing on Penn Treebank
1 code implementation • Findings of the Association for Computational Linguistics 2020 • Alireza Mohammadshahi, James Henderson
We propose the Graph2Graph Transformer architecture for conditioning on and predicting arbitrary graphs, and apply it to the challenging task of transition-based dependency parsing.
1 code implementation • EMNLP (WS) 2019 • Alireza Mohammadshahi, Remi Lebret, Karl Aberer
In this paper, we propose a new approach to learn multimodal multilingual embeddings for matching images and their relevant captions in two languages.