Research, Talks & Patents (Option B: Visual Modern)

My research focuses on multimodal foundation models, spatial acoustic reasoning, and generative audio architectures.

For complete citation indexes, visit my Google Scholar Profile and Google Research Profile.

Foundation Models & Publications

PhaseCoder: Microphone Geometry-Agnostic Spatial Audio Understanding for Multimodal LLMs
Artem Dementyev, Wazeer Zulfikar, Sinan Hersek, Pascal Getreuer, Anurag Kumar, Vivek Kumar
International Conference on Machine Learning (ICML), 2026
arXiv PDF

SpeechCompass: Enhancing Mobile Captioning with Diarization and Directional Guidance via Multi-Microphone Localization
Artem Dementyev, Wazeer Zulfikar, Sinan Hersek, Vivek Kumar
arXiv Preprint, 2025
arXiv PDF

SPAE: Semantic Pyramid AutoEncoder for Multimodal Generation with Frozen LLMs
Lijun Yu, Yong Cheng, Zhiruo Wang, Vivek Kumar, Wolfgang Macherey, Yanping Huang, David A. Ross, Irfan Essa, Yonatan Bisk, Ming-Hsuan Yang, Kevin P. Murphy, Alexander G. Hauptmann, Lu Jiang
Neural Information Processing Systems (NeurIPS), 2023
NeurIPS PDF

PyTorch DevCon

Detailed walkthrough of deep learning architectures applied to audio signals, comparing classical DSP against learned representations.