Zobrazeno 1 - 10
of 117
pro vyhledávání: '"Elidan, Gal"'
Large Language Models (LLM) technology is constantly improving towards human-like dialogue. Values are a basic driving force underlying human behavior, but little research has been done to study the values exhibited in text generated by LLMs. Here we
Externí odkaz:
http://arxiv.org/abs/2407.12878
Autor:
Caciularu, Avi, Jacovi, Alon, Ben-David, Eyal, Goldshtein, Sasha, Schuster, Tal, Herzig, Jonathan, Elidan, Gal, Globerson, Amir
Large Language Models (LLMs) often do not perform well on queries that require the aggregation of information across texts. To better evaluate this setting and facilitate modeling efforts, we introduce TACT - Text And Calculations through Tables, a d
Externí odkaz:
http://arxiv.org/abs/2406.03618
Autor:
Jurenka, Irina, Kunesch, Markus, McKee, Kevin R., Gillick, Daniel, Zhu, Shaojian, Wiltberger, Sara, Phal, Shubham Milind, Hermann, Katherine, Kasenberg, Daniel, Bhoopchand, Avishkar, Anand, Ankit, Pîslar, Miruna, Chan, Stephanie, Wang, Lisa, She, Jennifer, Mahmoudieh, Parsa, Rysbek, Aliya, Ko, Wei-Jen, Huber, Andrea, Wiltshire, Brett, Elidan, Gal, Rabin, Roni, Rubinovitz, Jasmin, Pitaru, Amit, McAllister, Mac, Wilkowski, Julia, Choi, David, Engelberg, Roee, Hackmon, Lidan, Levin, Adva, Griffin, Rachel, Sears, Michael, Bar, Filip, Mesar, Mia, Jabbour, Mana, Chaudhry, Arslan, Cohan, James, Thiagarajan, Sridhar, Levine, Nir, Brown, Ben, Gorur, Dilan, Grant, Svetlana, Hashimshoni, Rachel, Weidinger, Laura, Hu, Jieru, Chen, Dawn, Dolecki, Kuba, Akbulut, Canfer, Bileschi, Maxwell, Culp, Laura, Dong, Wen-Xin, Marchal, Nahema, Van Deman, Kelsie, Misra, Hema Bajaj, Duah, Michael, Ambar, Moran, Caciularu, Avi, Lefdal, Sandra, Summerfield, Chris, An, James, Kamienny, Pierre-Alexandre, Mohdi, Abhinit, Strinopoulous, Theofilos, Hale, Annie, Anderson, Wayne, Cobo, Luis C., Efron, Niv, Ananda, Muktha, Mohamed, Shakir, Heymans, Maureen, Ghahramani, Zoubin, Matias, Yossi, Gomes, Ben, Ibrahim, Lila
A major challenge facing the world is the provision of equitable and universal access to quality education. Recent advances in generative AI (gen AI) have created excitement about the potential of new technologies to offer a personal tutor for every
Externí odkaz:
http://arxiv.org/abs/2407.12687
Autor:
Rabin, Roni, Djerbetian, Alexandre, Engelberg, Roee, Hackmon, Lidan, Elidan, Gal, Tsarfaty, Reut, Globerson, Amir
Human communication often involves information gaps between the interlocutors. For example, in an educational dialogue, a student often provides an answer that is incomplete, and there is a gap between this answer and the perfect one expected by the
Externí odkaz:
http://arxiv.org/abs/2307.03319
Autor:
Roit, Paul, Ferret, Johan, Shani, Lior, Aharoni, Roee, Cideron, Geoffrey, Dadashi, Robert, Geist, Matthieu, Girgin, Sertan, Hussenot, Léonard, Keller, Orgad, Momchev, Nikola, Ramos, Sabela, Stanczyk, Piotr, Vieillard, Nino, Bachem, Olivier, Elidan, Gal, Hassidim, Avinatan, Pietquin, Olivier, Szpektor, Idan
Despite the seeming success of contemporary grounded text generation systems, they often tend to generate factually inconsistent text with respect to their input. This phenomenon is emphasized in tasks like summarization, in which the generated summa
Externí odkaz:
http://arxiv.org/abs/2306.00186
Autor:
Cohen, Deborah, Ryu, Moonkyung, Chow, Yinlam, Keller, Orgad, Greenberg, Ido, Hassidim, Avinatan, Fink, Michael, Matias, Yossi, Szpektor, Idan, Boutilier, Craig, Elidan, Gal
Despite recent advances in natural language understanding and generation, and decades of research on the development of conversational bots, building automated agents that can carry on rich open-ended conversations with humans "in the wild" remains a
Externí odkaz:
http://arxiv.org/abs/2208.02294
Supervised learning typically relies on manual annotation of the true labels. When there are many potential classes, searching for the best one can be prohibitive for a human annotator. On the other hand, comparing two candidate labels is often much
Externí odkaz:
http://arxiv.org/abs/2204.04670
Autor:
Nevo, Sella, Morin, Efrat, Rosenthal, Adi Gerzi, Metzger, Asher, Barshai, Chen, Weitzner, Dana, Voloshin, Dafi, Kratzert, Frederik, Elidan, Gal, Dror, Gideon, Begelman, Gregory, Nearing, Grey, Shalev, Guy, Noga, Hila, Shavitt, Ira, Yuklea, Liora, Royz, Moriah, Giladi, Niv, Levi, Nofar Peled, Reich, Ofir, Gilon, Oren, Maor, Ronnie, Timnat, Shahar, Shechter, Tal, Anisimov, Vladimir, Gigi, Yotam, Levin, Yuval, Moshe, Zach, Ben-Haim, Zvika, Hassidim, Avinatan, Matias, Yossi
The operational flood forecasting system by Google was developed to provide accurate real-time flood warnings to agencies and the public, with a focus on riverine floods in large, gauged rivers. It became operational in 2018 and has since expanded ge
Externí odkaz:
http://arxiv.org/abs/2111.02780
Autor:
Shoham, Yaron, Elidan, Gal
Despite seminal advances in reinforcement learning in recent years, many domains where the rewards are sparse, e.g. given only at task completion, remain quite challenging. In such cases, it can be beneficial to tackle the task both from its beginnin
Externí odkaz:
http://arxiv.org/abs/2105.01904
Autor:
Lang, Oran, Gandelsman, Yossi, Yarom, Michal, Wald, Yoav, Elidan, Gal, Hassidim, Avinatan, Freeman, William T., Isola, Phillip, Globerson, Amir, Irani, Michal, Mosseri, Inbar
Image classification models can depend on multiple different semantic attributes of the image. An explanation of the decision of the classifier needs to both discover and visualize these properties. Here we present StylEx, a method for doing this, by
Externí odkaz:
http://arxiv.org/abs/2104.13369