On the Utilization of Unique Node Identifiers in Graph Neural Networks

Autor: Bechler-Speicher, Maya, Eliasof, Moshe, Schönlieb, Carola-Bibiane, Gilad-Bachrach, Ran, Globerson, Amir

Graph Neural Networks have inherent representational limitations due to their message-passing structure. Recent work has suggested that these limitations can be overcome by using unique node identifiers (UIDs). Here we argue that despite the advantag

Externí odkaz: http://arxiv.org/abs/2411.02271

Real-time experiment-theory closed-loop interaction for autonomous materials science

Autor: Liang, Haotong, Wang, Chuangye, Yu, Heshan, Kirsch, Dylan, Pant, Rohit, McDannald, Austin, Kusne, A. Gilad, Zhao, Ji-Cheng, Takeuchi, Ichiro

Iterative cycles of theoretical prediction and experimental validation are the cornerstone of the modern scientific method. However, the proverbial "closing of the loop" in experiment-theory cycles in practice are usually ad hoc, often inherently dif

Externí odkaz: http://arxiv.org/abs/2410.17430

The Cost of Representation by Subset Repairs

Autor: Liu, Yuxi, Shen, Fangzhu, Ghosh, Kushagra, Gilad, Amir, Kimelfeld, Benny, Roy, Sudeepa

Datasets may include errors, and specifically violations of integrity constraints, for various reasons. Standard techniques for ``minimal-cost'' database repairing resolve these violations by aiming for minimum change in the data, and in the process,

Externí odkaz: http://arxiv.org/abs/2410.16501

The EMC Effect of Tritium and Helium-3 from the JLab MARATHON Experiment

Autor: Abrams, D., Albataineh, H., Aljawrneh, B. S., Alsalmi, S., Androic, D., Aniol, K., Armstrong, W., Arrington, J., Atac, H., Averett, T., Gayoso, C. Ayerbe, Bai, X., Bane, J., Barcus, S., Beck, A., Bellini, V., Bhatt, H., Bhetuwal, D., Biswas, D., Blyth, D., Boeglin, W., Bulumulla, D., Butler, J., Camsonne, A., Carmignotto, M., Castellanos, J., Chen, J. -P., Cloët, I. C., Cohen, E. O., Covrig, S., Craycraft, K., Cruz-Torres, R., Dongwi, B., Duran, B., Dutta, D., Fomin, N., Fuchey, E., Gal, C., Gautam, T. N., Gilad, S., Gnanvo, K., Gogami, T., Gomez, J., Gu, C., Habarakada, A., Hague, T., Hansen, J. -O., Hattawy, M., Hauenstein, F., Higinbotham, D. W., Holt, R. J., Hughes, E. W., Hyde, C., Ibrahim, H., Jian, S., Joosten, S., Karki, A., Karki, B., Katramatou, A. T., Keith, C., Keppel, C., Khachatryan, M., Khachatryan, V., Khanal, A., Kievsky, A., King, D., King, P. M., Korover, I., Kulagin, S. A., Kumar, K. S., Kutz, T., Lashley-Colthirst, N., Li, S., Li, W., Liu, H., Liuti, S., Liyanage, N., Markowitz, P., McClellan, R. E., Meekins, D., Beck, S. Mey-Tal, Meziani, Z. -E., Michaels, R., Mihovilovic, M., Nelyubin, V., Nguyen, D., Nuruzzaman, Nycz, M., Obrecht, R., Olson, M., Owen, V. F., Pace, E., Pandey, B., Pandey, V., Paolone, M., Papadopoulou, A., Park, S., Paul, S., Petratos, G. G., Petti, R., Piasetzky, E., Pomatsalyuk, R., Premathilake, S., Puckett, A. J. R., Punjabi, V., Ransome, R. D., Rashad, M. N. H., Reimer, P. E., Riordan, S., Roche, J., Salmè, G., Santiesteban, N., Sawatzky, B., Scopetta, S., Schmidt, A., Schmookler, B., Segal, J., Segarra, E. P., Shahinyan, A., Širca, S., Sparveris, N., Su, T., Suleiman, R., Szumila-Vance, H., Tadepalli, A. S., Tang, L., Tireman, W., Tortorici, F., Urciuoli, G. M., Wojtsekhowski, B., Wood, S., Ye, Z. H., Ye, Z. Y., Zhang, J.

Measurements of the EMC effect in the tritium and helium-3 mirror nuclei are reported. The data were obtained by the MARATHON Jefferson Lab experiment, which performed deep inelastic electron scattering from deuterium and the three-body nuclei, using

Externí odkaz: http://arxiv.org/abs/2410.12099

Are You Human? An Adversarial Benchmark to Expose LLMs

Autor: Gressel, Gilad, Pankajakshan, Rahul, Mirsky, Yisroel

Large Language Models (LLMs) have demonstrated an alarming ability to impersonate humans in conversation, raising concerns about their potential misuse in scams and deception. Humans have a right to know if they are conversing to an LLM. We evaluate

Externí odkaz: http://arxiv.org/abs/2410.09569