Scholar
Sarah Schwettmann
Google Scholar ID: FN2kigYAAAAJ
MIT
Citations & Impact
Citations
390
H-index
8
i10-index
8
Publications
20
Co-authors
8
list available
Publications
Resume (English only)
Academic Achievements
- ‘FIND: A Function Description Benchmark for Evaluating Interpretability Methods’, NeurIPS 2023
- ‘Multimodal Neurons in Pretrained Text-Only Transformers’, ICCV CVCL Workshop 2023 (Oral)
- ‘MILAN: Natural Language Descriptions of Deep Features’, ICLR 2022 (Oral)
- Work featured in MIT News articles such as ‘AI agents help explain other AI systems’ and ‘Demystifying machine-learning systems’
- Interviewed by MIT News in ‘3Q: The Interface Between Art and Neuroscience’
- Collaborated with The Metropolitan Museum of Art on ‘How Artificial Intelligence Sees Art History’
Background
- Research Scientist at MIT CSAIL with the MIT-IBM Watson AI Lab
- Research focuses on representations underlying intelligence in artificial (and previously biological) neural networks
- Broadly interested in human creativity in relating to the world—from the brain’s constructive role in perception to the explicit creation of experiential worlds in art
- Views this creative interface as a frontier for understanding intelligence and building intelligent machines
- Designed and co-taught MIT’s first course on Vision in Art and Neuroscience as a graduate student; continues teaching it every fall
