Zobrazeno 1 - 10
of 1 731
pro vyhledávání: '"Modarressi, A"'
Autor:
Kargaran, Amir Hossein, Modarressi, Ali, Nikeghbal, Nafiseh, Diesner, Jana, Yvon, François, Schütze, Hinrich
English-centric large language models (LLMs) often show strong multilingual capabilities. However, the multilingual performance of these models remains unclear and is not thoroughly evaluated for many languages. Most benchmarks for multilinguality fo
Externí odkaz:
http://arxiv.org/abs/2410.05873
Many datasets have been developed to train and evaluate document-level relation extraction (RE) models. Most of these are constructed using real-world data. It has been shown that RE models trained on real-world data suffer from factual biases. To ev
Externí odkaz:
http://arxiv.org/abs/2407.06699
While current large language models (LLMs) demonstrate some capabilities in knowledge-intensive tasks, they are limited by relying on their parameters as an implicit storage mechanism. As a result, they struggle with infrequent knowledge and temporal
Externí odkaz:
http://arxiv.org/abs/2404.11672
Video content has experienced a surge in popularity, asserting its dominance over internet traffic and Internet of Things (IoT) networks. Video compression has long been regarded as the primary means of efficiently managing the substantial multimedia
Externí odkaz:
http://arxiv.org/abs/2401.01163
Autor:
Modarressi, Ali, Fayyaz, Mohsen, Aghazadeh, Ehsan, Yaghoobzadeh, Yadollah, Pilehvar, Mohammad Taher
An emerging solution for explaining Transformer-based models is to use vector-based analysis on how the representations are formed. However, providing a faithful vector-based explanation for a multi-layer model could be challenging in three aspects:
Externí odkaz:
http://arxiv.org/abs/2306.02873
Large language models (LLMs) have significantly advanced the field of natural language processing (NLP) through their extensive parameters and comprehensive data utilization. However, existing LLMs lack a dedicated memory unit, limiting their ability
Externí odkaz:
http://arxiv.org/abs/2305.14322
Several proposals have been put forward in recent years for improving out-of-distribution (OOD) performance through mitigating dataset biases. A popular workaround is to train a robust model by re-weighting training examples based on a secondary bias
Externí odkaz:
http://arxiv.org/abs/2302.02852
Autor:
Fayyaz, Mohsen, Aghazadeh, Ehsan, Modarressi, Ali, Pilehvar, Mohammad Taher, Yaghoobzadeh, Yadollah, Kahou, Samira Ebrahimi
Current pre-trained language models rely on large datasets for achieving state-of-the-art performance. However, past research has shown that not all examples in a dataset are equally important during training. In fact, it is sometimes possible to pru
Externí odkaz:
http://arxiv.org/abs/2211.05610
Autor:
Taher Modarressi
Publikováno v:
American Journal of Preventive Cardiology, Vol 19, Iss , Pp 100706- (2024)
Background: Lipid-related risk and residual cardiovascular risk remain high in patients with type 2 diabetes (T2D) and atherosclerotic cardiovascular disease (ASCVD). Significant treatment gaps exist in implementation of pluripotent and effective the
Externí odkaz:
https://doaj.org/article/25893c841e63492f9957115178d9a9d8
There has been a growing interest in interpreting the underlying dynamics of Transformers. While self-attention patterns were initially deemed as the primary option, recent studies have shown that integrating other components can yield more accurate
Externí odkaz:
http://arxiv.org/abs/2205.03286