Zobrazeno 1 - 10
of 9 716
pro vyhledávání: '"P. -A. Michaud"'
Sparse autoencoders have recently produced dictionaries of high-dimensional vectors corresponding to the universe of concepts represented by large language models. We find that this concept universe has interesting structure at three levels: 1) The "
Externí odkaz:
http://arxiv.org/abs/2410.19750
Sparse autoencoders (SAEs) are a recent technique for decomposing neural network activations into human-interpretable features. However, in order for SAEs to identify all features represented in frontier models, it will be necessary to scale them up
Externí odkaz:
http://arxiv.org/abs/2410.08201
Autor:
Adderley, P. A., Ahmed, S., Allison, T., Bachimanchi, R., Baggett, K., BastaniNejad, M., Bevins, B., Bevins, M., Bickley, M., Bodenstein, R. M., Bogacz, S. A., Bruker, M., Burrill, A., Cardman, L., Creel, J., Chao, Y. -C., Cheng, G., Ciovati, G., Chattopadhyay, S., Clark, J., Clemens, W. A., Croke, G., Daly, E., Davis, G. K., Delayen, J., De Silva, S. U., Dickson, R., Diaz, M., Drury, M., Doolittle, L., Douglas, D., Feldl, E., Fischer, J., Freyberger, A., Ganni, V., Geng, R. L., Ginsburg, C., Gomez, J., Grames, J., Gubeli, J., Guo, J., Hannon, F., Hansknecht, J., Harwood, L., Henry, J., Hernandez-Garcia, C., Higgins, S., Higinbotham, D., Hofler, A. S., Hiatt, T., Hogan, J., Hovater, C., Hutton, A., Jones, C., Jordan, K., Joyce, M., Kazimi, R., Keesee, M., Kelley, M. J., Keppel, C., Kimber, A., King, L., Kjeldsen, P., Kneisel, P., Koval, J., Krafft, G. A., Lahti, G., Larrieu, T., Lauze, R., Leemann, C., Legg, R., Li, R., Lin, F., Machie, D., Mammosser, J., Macha, K., Mahoney, K., Marhauser, F., Mastracci, B., Matalevich, J., McCarter, J., McCaughan, M., Merminga, L., Michaud, R., Morozov, V., Mounts, C., Musson, J., Nelson, R., Oren, W., Overton, R. B., Palacios-Serrano, G., Park, H. -K., Phillips, L., Philip, S., Pilat, F., Plawski, T., Poelker, M., Powers, P., Powers, T., Preble, J., Reilly, T., Rimmer, R., Reece, C., Robertson, H., Roblin, Y., Rode, C., Satogata, T., Seidman, D. J., Seryi, A., Shabalina, A., Shin, I., Slominski, R., Slominski, C., Spata, M., Spell, D., Spradlin, J., Stirbet, M., Stutzman, M. L., Suhring, S., Surles-Law, K., Suleiman, R., Tennant, C., Tian, H., Turner, D., Tiefenback, M., Trofimova, O., Valente, A. -M., Wang, H., Wang, Y., White, K., Whitlatch, C., Whitlatch, T., Wiseman, M., Wissman, M. J., Wu, G., Yang, S., Yunn, B., Zhang, S., Zhang, Y.
Publikováno v:
Phys. Rev. Accel. Beams 27 (2024) 084802
This review paper describes the energy-upgraded CEBAF accelerator. This superconducting linac has achieved 12 GeV beam energy by adding 11 new high-performance cryomodules containing eighty-eight superconducting cavities that have operated CW at an a
Externí odkaz:
http://arxiv.org/abs/2408.16880
Autor:
Zanuz, Dante Colao, Ficheux, Quentin, Michaud, Laurent, Orekhov, Alexei, Hanke, Kilian, Flasby, Alexander, Panah, Mohsen Bahrami, Norris, Graham J., Kerschbaum, Michael, Remm, Ants, Swiadek, François, Hellings, Christoph, Lazăr, Stefania, Scarato, Colin, Lacroix, Nathan, Krinner, Sebastian, Eichler, Christopher, Wallraff, Andreas, Besse, Jean-Claude
The dominant contribution to the energy relaxation of state-of-the-art superconducting qubits is often attributed to their coupling to an ensemble of material defects which behave as two-level systems. These defects have varying microscopic character
Externí odkaz:
http://arxiv.org/abs/2407.18746
Autor:
Labbe, Mathieu, Michaud, François
Publikováno v:
IEEE/RSJ International Conference on Intelligent Robots and Systems, Chicago, IL, USA, 2014, pp. 2661-2666
For large-scale and long-term simultaneous localization and mapping (SLAM), a robot has to deal with unknown initial positioning caused by either the kidnapped robot problem or multi-session mapping. This paper addresses these problems by tying the S
Externí odkaz:
http://arxiv.org/abs/2407.15305
Autor:
Labbé, Mathieu, Michaud, François
Publikováno v:
IEEE Transactions on Robotics, vol. 29, no. 3, pp. 734-745, June 2013
In appearance-based localization and mapping, loop closure detection is the process used to determinate if the current observation comes from a previously visited location or a new one. As the size of the internal map increases, so does the time requ
Externí odkaz:
http://arxiv.org/abs/2407.15304
Autor:
Labbé, Mathieu, Michaud, François
Publikováno v:
IEEE/RSJ International Conference on Intelligent Robots and Systems, San Francisco, CA, USA, 2011, pp. 1271-1276
Loop closure detection is the process involved when trying to find a match between the current and a previously visited locations in SLAM. Over time, the amount of time required to process new observations increases with the size of the internal map,
Externí odkaz:
http://arxiv.org/abs/2407.15890
Autor:
Sguazzin, M., Jurado, B., Pibernat, J., Swartz, J. A., Grieser, M., Glorius, J., Litvinov, Yu. A., Berthelot, C., Włoch, B., Adamczewski-Musch, J., Alfaurt, P., Ascher, P., Audouin, L., Blank, B., Blaum, K., Brückner, B., Dellmann, S., Dillmann, I., Domingo-Pardo, C., Dupuis, M., Erbacher, P., Flayol, M., Forstner, O., Freire-Fernández, D., Gerbaux, M., Giovinazzo, J., Grévy, S., Griffin, C. J., Gumberidze, A., Heil, S., Heinz, A., Hess, R., Kurtulgil, D., Kurz, N., Leckenby, G., Litvinov, S., Lorentz, B., Méot, V., Michaud, J., Pérard, S., Petridis, N., Popp, U., Ramos, D., Reifarth, R., Roche, M., Sanjari, M. S., Sidhu, R. S., Spillmann, U., Steck, M., Stöhlker, Th., Thomas, B., Thulliez, L., Versteegen, M.
The probabilities for gamma-ray and particle emission as a function of the excitation energy of a decaying nucleus are valuable observables for constraining the ingredients of the models that describe the de-excitation of nuclei near the particle emi
Externí odkaz:
http://arxiv.org/abs/2407.14350
Autor:
SuperCDMS Collaboration, Albakry, M. F., Alkhatib, I., Alonso-González, D., Amaral, D. W. P., Anczarski, J., Aralis, T., Aramaki, T., Arnquist, I. J., Langroudy, I. Ataee, Azadbakht, E., Bathurst, C., Bhattacharyya, R., Biffl, A. J., Brink, P. L., Buchanan, M., Bunker, R., Cabrera, B., Calkins, R., Cameron, R. A., Cartaro, C., Cerdeño, D. G., Chang, Y. -Y., Chaudhuri, M., Chen, J. -H., Chen, R., Chott, N., Cooley, J., Coombes, H., Cushman, P., Cyna, R., Das, S., De Brienne, F., Dharani, S., di Vacri, M. L., Diamond, M. D., Elwan, M., Fascione, E., Figueroa-Feliciano, E., Fouts, K., Fritts, M., Germond, R., Ghaith, M., Golwala, S. R., Hall, J., Harms, S. A. S., Harris, K., Hassan, N., Hong, Z., Hoppe, E. W., Hsu, L., Huber, M. E., Iyer, V., Jardin, D., Kashyap, V. K. S., Keller, S. T. D., Kelsey, M. H., Kennard, K. T., Kubik, A., Kurinsky, N. A., Lee, M., Leyva, J., Liu, J., Liu, Y., Loer, B., Asamar, E. Lopez, Lukens, P., MacFarlane, D. B., Mahapatra, R., Mammo, J. S., Mast, N., Mayer, A. J., Theenhausen, H. Meyer zu, Michaud, É., Michielin, E., Mirabolfathi, N., Mirzakhani, M., Mohanty, B., Monteiro, D., Nelson, J., Neog, H., Novati, V., Orrell, J. L., Osborne, M. D., Oser, S. M., Pandey, L., Pandey, S., Partridge, R., Pedreros, D. S., Peng, W., Perna, L., Perry, W. L., Podviianiuk, R., Poudel, S. S., Pradeep, A., Pyle, M., Rau, W., Reid, E., Ren, R., Reynolds, T., Rios, M., Roberts, A., Robinson, A. E., Ryan, J. L., Saab, T., Sadek, D., Sadoulet, B., Sahoo, S. P., Saikia, I., Sander, J., Sattari, A., Schmidt, B., Schnee, R. W., Scorza, S., Serfass, B., Simchony, A., Sincavage, D. J., Sinervo, P., Street, J., Sun, H., Tanner, E., Terry, G. D., Toback, D., Verma, S., Villano, A. N., von Krosigk, B., Watkins, S. L., Wen, O., Williams, Z., Wilson, M. J., Winchell, J., Wykoff, K., Yellin, S., Young, B. A., Yu, T. C., Zatschler, B., Zatschler, S., Zaytsev, A., Zhang, E., Zheng, L., Zuniga, A., Zurowski, M. J.
This article presents constraints on dark-matter-electron interactions obtained from the first underground data-taking campaign with multiple SuperCDMS HVeV detectors operated in the same housing. An exposure of 7.63 g-days is used to set upper limit
Externí odkaz:
http://arxiv.org/abs/2407.08085
When a neural network can learn multiple distinct algorithms to solve a task, how does it "choose" between them during training? To approach this question, we take inspiration from ecology: when multiple species coexist, they eventually reach an equi
Externí odkaz:
http://arxiv.org/abs/2405.17420