Zobrazeno 1 - 10
of 7 172
pro vyhledávání: '"Seddik, A."'
Autor:
Pomaranski, David, Ito, Ryo, Tu, Ngoc Han, Ludwig, Arne, Wieck, Andreas D., Takada, Shintaro, Kaneko, Nobu-Hisa, Ouacel, Seddik, Bauerle, Christopher, Yamamoto, Michihisa
Standard approaches to quantum computing require significant overhead to correct for errors. The hardware size for conventional quantum processors in solids often increases linearly with the number of physical qubits, such as for transmon qubits in s
Externí odkaz:
http://arxiv.org/abs/2410.16244
In this paper, we introduce a method for fine-tuning Large Language Models (LLMs), inspired by Multi-Task learning in a federated manner. Our approach leverages the structure of each client's model and enables a learning scheme that considers other c
Externí odkaz:
http://arxiv.org/abs/2410.15524
Autor:
Firdoussi, Aymane El, Seddik, Mohamed El Amine, Hayou, Soufiane, Alami, Reda, Alzubaidi, Ahmed, Hacid, Hakim
Synthetic data has gained attention for training large language models, but poor-quality data can harm performance (see, e.g., Shumailov et al. (2023); Seddik et al. (2024)). A potential solution is data pruning, which retains only high-quality data
Externí odkaz:
http://arxiv.org/abs/2410.08942
Autor:
Alami, Reda, Almansoori, Ali Khalifa, Alzubaidi, Ahmed, Seddik, Mohamed El Amine, Farooq, Mugariya, Hacid, Hakim
We demonstrate that preference optimization methods can effectively enhance LLM safety. Applying various alignment techniques to the Falcon 11B model using safety datasets, we achieve a significant boost in global safety score (from $57.64\%$ to $99.
Externí odkaz:
http://arxiv.org/abs/2409.07772
Autor:
Shaju, Jashwanth, Pavlovska, Elina, Suba, Ralfs, Wang, Junliang, Ouacel, Seddik, Vasselon, Thomas, Aluffi, Matteo, Mazzella, Lucas, Geffroy, Clement, Ludwig, Arne, Wieck, Andreas D., Urdampiletta, Matias, Bäuerle, Christopher, Kashcheyevs, Vyacheslavs, Sellier, Hermann
Emergence of universal collective behaviour from interactions in a sufficiently large group of elementary constituents is a fundamental scientific paradigm. In physics, correlations in fluctuating microscopic observables can provide key information a
Externí odkaz:
http://arxiv.org/abs/2408.14458
Autor:
Ouacel, Seddik, Mazzella, Lucas, Kloss, Thomas, Aluffi, Matteo, Vasselon, Thomas, Edlbauer, Hermann, Wang, Junliang, Geffroy, Clement, Shaju, Jashwanth, Yamamoto, Michihisa, Pomaranski, David, Takada, Shintaro, Kaneko, Nobu-Hisa, Georgiou, Giorgos, Waintal, Xavier, Urdampilleta, Matias, Ludwig, Arne, Wieck, Andreas D., Sellier, Hermann, Bäuerle, Christopher
Electronic flying qubits offer an interesting alternative to photonic qubits: electrons propagate slower, hence easier to control in real time, and Coulomb interaction enables direct entanglement between different qubits. While their coherence time i
Externí odkaz:
http://arxiv.org/abs/2408.13025
Autor:
Malartic, Quentin, Chowdhury, Nilabhra Roy, Cojocaru, Ruxandra, Farooq, Mugariya, Campesan, Giulia, Djilali, Yasser Abdelaziz Dahou, Narayan, Sanath, Singh, Ankit, Velikanov, Maksim, Boussaha, Basma El Amel, Al-Yafeai, Mohammed, Alobeidli, Hamza, Qadi, Leen Al, Seddik, Mohamed El Amine, Fedyanin, Kirill, Alami, Reda, Hacid, Hakim
We introduce Falcon2-11B, a foundation model trained on over five trillion tokens, and its multimodal counterpart, Falcon2-11B-vlm, which is a vision-to-text model. We report our findings during the training of the Falcon2-11B which follows a multi-s
Externí odkaz:
http://arxiv.org/abs/2407.14885
Autor:
Elbakary, Ahmed, Issaid, Chaouki Ben, Shehab, Mohammad, Seddik, Karim, ElBatt, Tamer, Bennis, Mehdi
Federated learning is a machine learning approach where multiple devices collaboratively learn with the help of a parameter server by sharing only their local updates. While gradient-based optimization techniques are widely adopted in this domain, th
Externí odkaz:
http://arxiv.org/abs/2406.06655
Non-orthogonal multiple access (NOMA) is widely recognized for its spectral and energy efficiency, which allows more users to share the network resources more effectively. This paper provides a generalized bit error rate (BER) performance analysis of
Externí odkaz:
http://arxiv.org/abs/2405.19639
This paper provides theoretical insights into high-dimensional binary classification with class-conditional noisy labels. Specifically, we study the behavior of a linear classifier with a label noisiness aware loss function, when both the dimension o
Externí odkaz:
http://arxiv.org/abs/2405.14088