We present a series of two studies conducted to understand user’s affective states during voice-based human-machine interactions. Emphasis is placed on the cases of communication errors or failures.
In this paper, we investigate uncovering the unique socio-cultural perspectives embedded in human-made art, which in turn, can be valuable in expanding the horizon of AI ethics.
Error maps can indicate when a correctly attended region may be processed incorrectly leading to an incorrect answer, and hence, improve users’ understanding of those cases.
We identify three critical biases stemming from the question-answer generation process and memorization capabilities of large deep models.
Senior Technical Director, Vision and Learning Laboratory, Center for Vision Technologies
Our experiments focus on zero-shot question answering, using the taxonomy to provide proximal context that helps the model answer questions by being relevant to those questions.
While literature has demonstrated great successes via representation learning, in this work, we show that improvement of downstream tasks can also be achieved by appropriate designs of the adaptation process.
We present novel calibration solutions via domain generalization. Our core idea is to leverage multiple calibration domains to reduce the effective distribution disparity between the target and calibration domains for improved calibration transfer without needing any data from the target domain.
We introduce Hybrid Consistency Training to jointly leverage interpolation consistency, including interpolating hidden features, that imposes linear behavior locally and data augmentation consistency that learns robust embeddings against sample variations.