Part-Of-Speech Sensitivity of Routers in Mixture of Experts Models

Autor: Antoine, Elie, Béchet, Frédéric, Langlais, Philippe

This study investigates the behavior of model-integrated routers in Mixture of Experts (MoE) models, focusing on how tokens are routed based on their linguistic features, specifically Part-of-Speech (POS) tags. The goal is to explore across different

Externí odkaz: http://arxiv.org/abs/2412.16971

Demonstrating dynamic surface codes

Autor: Eickbusch, Alec, McEwen, Matt, Sivak, Volodymyr, Bourassa, Alexandre, Atalaya, Juan, Claes, Jahan, Kafri, Dvir, Gidney, Craig, Warren, Christopher W., Gross, Jonathan, Opremcak, Alex, Miao, Nicholas Zobrist Kevin C., Roberts, Gabrielle, Satzinger, Kevin J., Bengtsson, Andreas, Neeley, Matthew, Livingston, William P., Greene, Alex, Rajeev, Acharya, Beni, Laleh Aghababaie, Aigeldinger, Georg, Alcaraz, Ross, Andersen, Trond I., Ansmann, Markus, Frank, Arute, Arya, Kunal, Asfaw, Abraham, Babbush, Ryan, Ballard, Brian, Bardin, Joseph C., Bilmes, Alexander, Jenna, Bovaird, Bowers, Dylan, Brill, Leon, Broughton, Michael, Browne, David A., Buchea, Brett, Buckley, Bob B., Tim, Burger, Burkett, Brian, Bushnell, Nicholas, Cabrera, Anthony, Campero, Juan, Chang, Hung-Shen, Chiaro, Ben, Chih, Liang-Ying, Cleland, Agnetta Y., Cogan, Josh, Collins, Roberto, Conner, Paul, Courtney, William, Alexander, Crook, L., Curtin, Ben, Das, Sayan, Barba, Alexander Del Toro, Demura, Sean, De Lorenzo, Laura, Di Paolo, Agustin, Donohoe, Paul, Drozdov, Ilya K., Dunsworth, Andrew, Elbag, Aviv Moshe, Elzouka, Mahmoud, Erickson, Catherine, Ferreira, Vinicius S., Burgos, Leslie Flores, Forati, Ebrahim, Fowler, Austin G., Foxen, Brooks, Ganjam, Suhas, Gonzalo, Garcia, Gasca, Robert, Genois, Élie, Giang, William, Gilboa, Dar, Gosula, Raja, Dau, Alejandro Grajales, Dietrich, Graumann, Ha, Tan, Habegger, Steve, Hansen, Monica, Harrigan, Matthew P., Harrington, Sean D., Heslin, Stephen, Heu, Paula, Higgott, Oscar, Hiltermann, Reno, Hilton, Jeremy, Huang, Hsin-Yuan, Huff, Ashley, Huggins, William J., Jeffrey, Evan, Jiang, Zhang, Jin, Xiaoxuan, Jones, Cody, Joshi, Chaitali, Juhas, Pavol, Kabel, Andreas, Kang, Hui, Amir, Karamlou, H., Kechedzhi, Kostyantyn, Khaire, Trupti, Khattar, Tanuj, Khezri, Mostafa, Kim, Seon, Kobrin, Bryce, Korotkov, Alexander N., Kostritsa, Fedor, Kreikebaum, John Mark, Kurilovich, Vladislav D., Landhuis, David, Tiano, Lange-Dei, Langley, Brandon W., Lau, Kim-Ming, Ledford, Justin, Lee, Kenny, Lester, Brian J., Guevel, Loïck Le, Wing, Li, Yan, Lill, Alexander T., Locharla, Aditya, Lucero, Erik, Lundahl, Daniel, Lunt, Aaron, Madhuk, Sid, Maloney, Ashley, Mandrà, Salvatore, Martin, Leigh S., Martin, Orion, Maxfield, Cameron, McClean, Jarrod R., Meeks, Seneca, Anthony, Megrant, Molavi, Reza, Molina, Sebastian, Montazeri, Shirin, Movassagh, Ramis, Newman, Michael, Nguyen, Anthony, Nguyen, Murray, Ni, Chia-Hung, Oas, Logan, Orosco, Raymond, Ottosson, Kristoffer, Pizzuto, Alex, Potter, Rebecca, Pritchard, Orion, Quintana, Chris, Ramachandran, Ganesh, Reagor, Matthew J., Rhodes, David M., Rosenberg, Eliott, Rossi, Elizabeth, Sankaragomathi, Kannan, Schurkus, Henry F., Shearn, Michael J., Shorter, Aaron, Shutty, Noah, Shvarts, Vladimir, Small, Spencer, Smith, W. Clarke, Springer, Sofia, Sterling, George, Suchard, Jordan, Szasz, Aaron, Sztein, Alex, Thor, Douglas, Tomita, Eifu, Torres, Alfredo, Torunbalci, M. Mert, Vaishnav, Abeer, Vargas, Justin, Sergey, Vdovichev, Vidal, Guifre, Heidweiller, Catherine Vollgraff, Waltman, Steven, Waltz, Jonathan, Wang, Shannon X., Ware, Brayden, Weidel, Travis, White, Theodore, Wong, Kristi, Woo, Bryan W. K., Woodson, Maddy, Xing, Cheng, Yao, Z. Jamie, Yeh, Ping, Ying, Bicheng, Yoo, Juhwan, Yosri, Noureldin, Young, Grayson, Zalcman, Adam, Yaxing, Zhang, Zhu, Ningfeng, Boixo, Sergio, Kelly, Julian, Smelyanskiy, Vadim, Neven, Hartmut, Bacon, Dave, Chen, Zijun, Klimov, Paul V., Roushan, Pedram, Neill, Charles, Chen, Yu, Morvan, Alexis

A remarkable characteristic of quantum computing is the potential for reliable computation despite faulty qubits. This can be achieved through quantum error correction, which is typically implemented by repeatedly applying static syndrome checks, per

Externí odkaz: http://arxiv.org/abs/2412.14360

Multi-Task Reinforcement Learning for Quadrotors

Autor: Xing, Jiaxu, Geles, Ismail, Song, Yunlong, Aljalbout, Elie, Scaramuzza, Davide

Publikováno v: Robotics and Automation Letters 2024

Reinforcement learning (RL) has shown great effectiveness in quadrotor control, enabling specialized policies to develop even human-champion-level performance in single-task scenarios. However, these specialized policies often struggle with novel tas

Externí odkaz: http://arxiv.org/abs/2412.12442

GEM: A Generalizable Ego-Vision Multimodal World Model for Fine-Grained Ego-Motion, Object Dynamics, and Scene Composition Control

Autor: Hassan, Mariam, Stapf, Sebastian, Rahimi, Ahmad, Rezende, Pedro M B, Haghighi, Yasaman, Brüggemann, David, Katircioglu, Isinsu, Zhang, Lin, Chen, Xiaoran, Saha, Suman, Cannici, Marco, Aljalbout, Elie, Ye, Botao, Wang, Xi, Davtyan, Aram, Salzmann, Mathieu, Scaramuzza, Davide, Pollefeys, Marc, Favaro, Paolo, Alahi, Alexandre

We present GEM, a Generalizable Ego-vision Multimodal world model that predicts future frames using a reference frame, sparse features, human poses, and ego-trajectories. Hence, our model has precise control over object dynamics, ego-agent motion and

Externí odkaz: http://arxiv.org/abs/2412.11198

On the Significance of Intermediate Latents: Distinguishing Quantum Causal Scenarios with Indistinguishable Classical Analogs

Autor: Centeno, Daniel, Wolfe, Elie

The use of graphical models to represent causal hypotheses has enabled revolutionary progress in the study of the foundations of quantum theory. Here we consider directed acyclic graphs each of which contains both nodes representing observed variable

Externí odkaz: http://arxiv.org/abs/2412.10238