talks and posters in conferences/seminars

invited talks

  1. Building Trustworthy Systems: Compositional Adversarial Robustness and Low-Voltage Inference, TrustML Workshop @ UBC, Jun. 23, 2023 (Vancouver, BC Canada) [link]
  2. Bit Errors of SRAM-Based Weight Storage: Trade-offs Between Energy and Accuracy, AI Accelerators Short Course, Taiwan AI Academy, Mar. 29, 2022 (invited by Prof. H. T. Kung) (Hsinchu, Taiwan)

contributed talks

  1. NCTV: Neural Clamping Toolkit and Visualization for Neural Network Calibration, AAAI 2023 Demo, Feb., 2023 (Washington, DC)
  2. CARBEN: Composite Adversarial Robustness Benchmark, NeurIPS 2022 IBM’s Expo, Nov., 2022 (New Orleans, LA)
  3. CARBEN: Composite Adversarial Robustness Benchmark, IJCAI 2022 Demo, Jul., 2022 (Vienna, Austria)


  1. Towards Compositional Adversarial Robustness: Generalizing Adversarial Training to Composite Semantic Perturbations, CVPR 2023, Jun., 2023 (Vancouver, BC Canada)
  2. NeuralFuse: Improving the Accuracy of Access-Limited Neural Network Inference in Low-Voltage Regimes, NeurIPS 2022 MLSys Workshop, Dec., 2022 (New Orleans, LA)