Paper

[1] Soham Dan, Xinran Han, Dan Roth, “Compositional Data and Task Augmentation for Instruction Following” (under submission)

[2] Soham Dan, Michael Zhou, Dan Roth, “Generalization in Instruction Following Systems”, NAACL 2021

[3] Kaifu Wang, Qiang Ning, Dan Roth, Learnability with Indirect Supervision Signals NeurIPS 2020

[4] Sangdon Park, Shuo Li, Insup Lee, Osbert Bastani. PAC Confidence Predictions for Deep Neural Network Classifiers. In ICLR, 2021

[5] Jorge Mendez and Eric Eaton. Lifelong Learning of Compositional Structures. To appear in the International Conference on Learning Representations (ICLR). https://openreview.net/pdf?id=ADWd4TJO13G

[6] Y. Kantaros, T. Carpenter, K. Sridhar, I. Lee, J. Weimer: `Real-Time Detectors for Adversarial Digital and Physical Inputs to Perception Systems', 12th ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS), 2021

[7] Jang, Sooyong, Radoslav Ivanov, Insup Lee, and James Weimer. "Confidence Calibration with Bounded Error Using Transformations." arXiv preprint arXiv:2102.12680 (2021)

[8] Jang, Sooyong, Insup Lee, and James Weimer. "Improving Classifier Confidence using Lossy Label-Invariant Transformations." arXiv preprint arXiv:2011.04182 (2020)

[9] Kantaros, Yiannis, Taylor Carpenter, Sangdon Park, Radoslav Ivanov, Sooyong Jang, Insup Lee, and James Weimer. "VisionGuard: Runtime detection of adversarial inputs to perception systems." arXiv preprint arXiv:2002.09792 (2020)