Hao Dong at NeurIPS 2024
Hao's work on MultiOOD: Scaling Out-of-Distribution Detection for Multiple Modalities has been accepted by hashtag#NeurIPS2024 as a π¦π½πΌππΉπΆπ΄π΅π!
Do not miss the opportunity to follow Hao Dong's amazing work in the upcoming ECCV and NeurIPS conferences. Hao is a joint member of the IMOS Lab - EPFL (under supervision of Olga Fink) and Chair of Structural Mechanics and Monitoring at ETH Zurich (under supervision of Eleni Chatzi).
π‘ His work on MultiOOD: Scaling Out-of-Distribution Detection for Multiple Modalities has been accepted by NeurIPS2024 as a π¦π½πΌππΉπΆπ΄π΅π!
This work, in collaboration with Yue Zhao, introduces a first-of-its-kind benchmark for Multimodal OOD Detection, highlighting the significance of integrating more modalities and introducing an Agree-to-Disagree (A2D) training algorithm, inspired by the observation of the Modality Prediction Discrepancy phenomenon.
external page paper | external page code
π‘ Prior to this, you can catch him at hashtag#ECCV2024 in Milan next week presenting "Towards Multimodal Open-Set Domain Generalization and Adaptation through Self-supervision"
external page paper | external page code
Hao's work on Multimodal Learning and Anomaly detection is funded as part of the ETH Mobility project SENTINEL, of the ETH Mobility Initiative, administered by the Centre for Sustainable Future Mobility - ETH Zurich