Schnelle Erfolge in den ersten 100 Tagen erzielen.

Autor: Leemann, Niklaus

Publikováno v: Changement. winter2024, Issue 9, p66-69. 4p.

Auto-GDA: Automatic Domain Adaptation for Efficient Grounding Verification in Retrieval Augmented Generation

Autor: Leemann, Tobias, Petridis, Periklis, Vietri, Giuseppe, Manousakas, Dionysis, Roth, Aaron, Aydore, Sergul

While retrieval augmented generation (RAG) has been shown to enhance factuality of large language model (LLM) outputs, LLMs still suffer from hallucination, generating incorrect or irrelevant information. One common detection strategy involves prompt

Externí odkaz: http://arxiv.org/abs/2410.03461

The Continuous Electron Beam Accelerator Facility at 12 GeV

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

Bundesgericht, I. zivilrechtliche Abteilung, 4A_2/2023, Urteil vom 6. Oktober 2023, A. gegen FC B., Bundesrichterin Kiss, präsidierendes Mitglied, Bundesrichterin Hohl, Bundesrichter Rüedi, Gerichtsschreiber Leemann.

Publikováno v: ASA Bulletin. Dec2023, Vol. 41 Issue 4, p929-939. 11p.

Attention Mechanisms Don't Learn Additive Models: Rethinking Feature Importance for Transformers

Autor: Leemann, Tobias, Fastowski, Alina, Pfeiffer, Felix, Kasneci, Gjergji

We address the critical challenge of applying feature attribution methods to the transformer architecture, which dominates current applications in natural language processing and beyond. Traditional attribution methods to explainable AI (XAI) explici

Externí odkaz: http://arxiv.org/abs/2405.13536