Selected:
Method prevents an AI model from being overconfident about wrong answers
September 16, 2024
More efficient than other approaches, the “Thermometer” technique could help someone know when they should trust a large language model.
Reasoning and reliability in AI
January 22, 2024
PhD students interning with the MIT-IBM Watson AI Lab look to improve natural language usage.
Efficient technique improves machine-learning models’ reliability
February 16, 2023
The method enables a model to determine its confidence in a prediction, while using no additional data and far fewer computing resources than other methods.