IBM Research | ExBERT: A Visual Tool to Explore BERT @ibmresearch | Uploaded June 2020 | Updated October 2024, 4 days ago.
Learn how to uncover insights into what deep Transformer models understand about human language by interactively exploring their learned attentions and contextual embeddings.
exbert.net

The Short: Honoring laser pioneers, AI algorithms that mimic the brain, Donuts in hyperspace!

Sustainable materials, sustainable products, sustainable planet

The Short: Recapping all of the news and announcements from THINK 2024!

IBM Research - Israel: 50 years of creating a better future

ICLR Paper: Learn Step Size Quantization

In-memory physical superposition meets few-shot continual learning

The Short: AI chips come to UAlbany, protecting AI models from attack, and tiny magnetic molecules

Capturing and transforming CO2 to mitigate climate change

$Whats Next in AI : AI We Can Trust Tune in to our multi-stream event spanning three weeks in November and featuring the AI scientists developing AI technologies for real-world implementation. Leaders agree Artificial Intelligence technologies are key to competitive advantage in the 21stCentury, but only a fraction of organizations are successfully executing in AI. In this virtual event, scientists and business experts from the MIT-IBM Watson AI Lab come together to review three key barriers to adoption –trust, scalability, and reasoning –and how we can solve these challenges through scientific advancement. 11/05 AI We Can Trust 11/12 AI We Can Scale 11/19 AI We Can Reason With Here are the details from the discussion in this video 9:05 Welcome and Introduction Mark Weber 9:15 AI Science for Real-World Impact Aude Oliva 9:30 Business Discussion Mark Weber, Aude Oliva, David Cox 9:45 Certifying Robustness Lily Weng 10:00 Business Discussion Mark Weber, Lily Weng, and David Cox 10:15 Safe AI Armando Solar-Lezama 10:30 Business Discussion Mark Weber and Armando Solar-Lezama 10:45 Algorithmic Fairness Mikhail Yurochkin 11:00 Business Discussion Mark Weber, Mikhail Yurochkin, and Sherif Botros 11:15 Identifying unreliable predictions in clinical risk models Collin Stultz 11:30 Business Discussion Mark Weber, Collin Stultz, Jianying Hu and Ron Lancaster 11:45 Closing thoughts and preview of the next session Mark Weber, Aude Oliva, David Cox$

Understanding the NIST standards and IBMs contributions to post-quantum cryptography

ExBERT: A Visual Tool to Explore BERT @ibmresearch