Zobrazeno 1 - 6
of 6
pro vyhledávání: '"Wedin, Ben"'
Large language models (LLMs) are highly capable at a variety of tasks given the right prompt, but writing one is still a difficult and tedious process. In this work, we introduce ConstitutionalExperts, a method for learning a prompt consisting of con
Externí odkaz:
http://arxiv.org/abs/2403.04894
Autor:
Petridis, Savvas, Wedin, Ben, Wexler, James, Donsbach, Aaron, Pushkarna, Mahima, Goyal, Nitesh, Cai, Carrie J., Terry, Michael
Large language model (LLM) prompting is a promising new approach for users to create and customize their own chatbots. However, current methods for steering a chatbot's outputs, such as prompt engineering and fine-tuning, do not support users in conv
Externí odkaz:
http://arxiv.org/abs/2310.15428
Traditional recommender systems leverage users' item preference history to recommend novel content that users may like. However, modern dialog interfaces that allow users to express language-based preferences offer a fundamentally different modality
Externí odkaz:
http://arxiv.org/abs/2307.14225
Natural interaction with recommendation and personalized search systems has received tremendous attention in recent years. We focus on the challenge of supporting people's understanding and control of these systems and explore a fundamentally new way
Externí odkaz:
http://arxiv.org/abs/2205.09403
Developing a suitable Deep Neural Network (DNN) often requires significant iteration, where different model versions are evaluated and compared. While metrics such as accuracy are a powerful means to succinctly describe a model's performance across a
Externí odkaz:
http://arxiv.org/abs/2201.11196
Autor:
Kapishnikov, Andrei, Venugopalan, Subhashini, Avci, Besim, Wedin, Ben, Terry, Michael, Bolukbasi, Tolga
Publikováno v:
Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021, pp. 5050-5058
Integrated Gradients (IG) is a commonly used feature attribution method for deep neural networks. While IG has many desirable properties, the method often produces spurious/noisy pixel attributions in regions that are not related to the predicted cla
Externí odkaz:
http://arxiv.org/abs/2106.09788