1 code implementation • 30 Apr 2024 • Alireza Salemi, Hamed Zamani
This lays the groundwork for us to build a large-scale experimentation ecosystem consisting of 18 RAG systems that engage in training and 18 unknown RAG systems that use the uRAG as the new users of the search engine.
1 code implementation • 21 Apr 2024 • Alireza Salemi, Hamed Zamani
Furthermore, evaluation of the retrieval model's performance based on query-document relevance labels shows a small correlation with the RAG system's downstream performance.
1 code implementation • 9 Apr 2024 • Alireza Salemi, Surya Kallumadi, Hamed Zamani
This paper studies retrieval-augmented approaches for personalizing large language models (LLMs), which potentially have a substantial impact on various applications and domains.
1 code implementation • 28 Jun 2023 • Alireza Salemi, Mahta Rafiee, Hamed Zamani
The proposed approach leads to 26. 9% Precision@5 improvements compared to the current state-of-the-art asymmetric architecture.
1 code implementation • 26 Apr 2023 • Alireza Salemi, Juan Altmayer Pizzorno, Hamed Zamani
Utilizing the passages retrieved by DEDR, we further introduce MM-FiD, an encoder-decoder multi-modal fusion-in-decoder model, for generating a textual answer for KI-VQA tasks.
1 code implementation • 22 Apr 2023 • Alireza Salemi, Sheshera Mysore, Michael Bendersky, Hamed Zamani
This paper highlights the importance of personalization in large language models and introduces the LaMP benchmark -- a novel benchmark for training and evaluating language models for producing personalized outputs.
1 code implementation • 3 Apr 2023 • Alireza Salemi, Amirhossein Abaskohi, Sara Tavakoli, Yadollah Yaghoobzadeh, Azadeh Shakery
Multilingual pre-training on monolingual data ignores the availability of parallel data in many language pairs.
1 code implementation • EMNLP 2021 • Alireza Salemi, Emad Kebriaei, Ghazal Neisi Minaei, Azadeh Shakery
Current pre-training works in abstractive summarization give more points to the summaries with more words in common with the main text and pay less attention to the semantic similarity between generated sentences and the original document.
1 code implementation • SEMEVAL 2021 • Alireza Salemi, Nazanin Sabri, Emad Kebriaei, Behnam Bahrak, Azadeh Shakery
Detecting which parts of a sentence contribute to that sentence's toxicity -- rather than providing a sentence-level verdict of hatefulness -- would increase the interpretability of models and allow human moderators to better understand the outputs of the system.