Zobrazeno 1 - 10
of 796
pro vyhledávání: '"Hénaff, P."'
Autor:
Klissarov, Martin, Henaff, Mikael, Raileanu, Roberta, Sodhani, Shagun, Vincent, Pascal, Zhang, Amy, Bacon, Pierre-Luc, Precup, Doina, Machado, Marlos C., D'Oro, Pierluca
Describing skills in natural language has the potential to provide an accessible way to inject human knowledge about decision-making into an AI system. We present MaestroMotif, a method for AI-assisted skill design, which yields high-performing and a
Externí odkaz:
http://arxiv.org/abs/2412.08542
Autor:
Udandarao, Vishaal, Parthasarathy, Nikhil, Naeem, Muhammad Ferjad, Evans, Talfan, Albanie, Samuel, Tombari, Federico, Xian, Yongqin, Tonioni, Alessio, Hénaff, Olivier J.
Knowledge distillation (KD) is the de facto standard for compressing large-scale models into smaller ones. Prior works have explored ever more complex KD strategies involving different objective functions, teacher-ensembles, and weight inheritance. I
Externí odkaz:
http://arxiv.org/abs/2411.18674
Large-scale multimodal representation learning successfully optimizes for zero-shot transfer at test time. Yet the standard pretraining paradigm (contrastive learning on large amounts of image-text data) does not explicitly encourage representations
Externí odkaz:
http://arxiv.org/abs/2411.15099
Automatically synthesizing dense rewards from natural language descriptions is a promising paradigm in reinforcement learning (RL), with applications to sparse reward problems, open-ended exploration, and hierarchical skill design. Recent works have
Externí odkaz:
http://arxiv.org/abs/2410.23022
Autor:
Faou, Erwan, Henaff, Yoann Le
In this paper we generalize the spectral concentration problem as formulated by Slepian, Pollak and Landau in the 1960s. We show that a generalized version with arbitrary space and Fourier masks is well-posed, and we prove some new results concerning
Externí odkaz:
http://arxiv.org/abs/2410.01465
Autor:
Roth, Karsten, Udandarao, Vishaal, Dziadzio, Sebastian, Prabhu, Ameya, Cherti, Mehdi, Vinyals, Oriol, Hénaff, Olivier, Albanie, Samuel, Bethge, Matthias, Akata, Zeynep
Multimodal foundation models serve numerous applications at the intersection of vision and language. Still, despite being pretrained on extensive data, they become outdated over time. To keep models updated, research into continual pretraining mainly
Externí odkaz:
http://arxiv.org/abs/2408.14471
Autor:
Abreu, Y., Amhis, Y., Arnold, L., Beaumont, W., Bolognino, I., Bongrand, M., Boursette, D., Buridon, V., Chanal, H., Coupé, B., Crochet, P., Cussans, D., D'Hondt, J., Durand, D., Fallot, M., Galbinski, D., Gallego, S., Ghys, L., Giot, L., Graves, K., Guillon, B., Hayashida, S., Henaff, D., Hosseini, B., Kalcheva, S., Kalousis, L. N., Keloth, R., Koch, L., Labare, M., Lehaut, G., Manley, S., Manzanillas, L., Mermans, J., Michiels, I., Monteil, S., Moortgat, C., Newbold, D., Pestel, V., Petridis, K., Piñera, I., de Roeck, A., Roy, N., Ryckbosch, D., Ryder, N., Saunders, D., Schune, M. H., Settimo, M., Sfar, H. Rejeb, Simard, L., Vacheret, A., Van Dyck, S., Van Mulders, P., Van Remortel, N., Vandierendonck, G., Vercaemer, S., Verstraeten, M., Viaud, B., Weber, A., Yeresko, M., Yermia, F.
In this letter we report the first scientific result based on antineutrinos emitted from the BR2 reactor at SCK CEN. The SoLid experiment uses a novel type of highly granular detector whose basic detection unit combines two scintillators, PVT and 6Li
Externí odkaz:
http://arxiv.org/abs/2407.14382
Autor:
Bandic, Medina, Henaff, Pablo le, Ovide, Anabel, Escofet, Pau, Rached, Sahar Ben, Rodrigo, Santiago, van Someren, Hans, Abadal, Sergi, Alarcon, Eduard, Almudever, Carmen G., Feld, Sebastian
Application-specific quantum computers offer the most efficient means to tackle problems intractable by classical computers. Realizing these architectures necessitates a deep understanding of quantum circuit properties and their relationship to execu
Externí odkaz:
http://arxiv.org/abs/2407.12640
Autor:
Beyer, Lucas, Steiner, Andreas, Pinto, André Susano, Kolesnikov, Alexander, Wang, Xiao, Salz, Daniel, Neumann, Maxim, Alabdulmohsin, Ibrahim, Tschannen, Michael, Bugliarello, Emanuele, Unterthiner, Thomas, Keysers, Daniel, Koppula, Skanda, Liu, Fangyu, Grycner, Adam, Gritsenko, Alexey, Houlsby, Neil, Kumar, Manoj, Rong, Keran, Eisenschlos, Julian, Kabra, Rishabh, Bauer, Matthias, Bošnjak, Matko, Chen, Xi, Minderer, Matthias, Voigtlaender, Paul, Bica, Ioana, Balazevic, Ivana, Puigcerver, Joan, Papalampidi, Pinelopi, Henaff, Olivier, Xiong, Xi, Soricut, Radu, Harmsen, Jeremiah, Zhai, Xiaohua
PaliGemma is an open Vision-Language Model (VLM) that is based on the SigLIP-So400m vision encoder and the Gemma-2B language model. It is trained to be a versatile and broadly knowledgeable base model that is effective to transfer. It achieves strong
Externí odkaz:
http://arxiv.org/abs/2407.07726
Data curation is an essential component of large-scale pretraining. In this work, we demonstrate that jointly selecting batches of data is more effective for learning than selecting examples independently. Multimodal contrastive objectives expose the
Externí odkaz:
http://arxiv.org/abs/2406.17711