Research Impact

Ali Vosoughi • University of Rochester | Dissertation complete • alivosoughi.com

Publications & Citations

• First-author publications: IEEE TMM, ICASSP, EUSIPCO
• Survey leadership: Video Understanding with LLMs (IEEE TCSVT 2025)

Industry Research Collaborations

• Apple (2026): machine learning internship, a multi-agent multimodal system
• Microsoft Research (2024): Unified audiovisual encoder → EUSIPCO 2025
• Bosch AI Research (2023): Counterfactual audio-language → ICASSP 2024 + US patent application
• Smule AI (2025): Spatial audio generation for real-time applications
• Clinical AI research collaborations: Aidoc, Carestream, and Gleamer (radiology AI studies with the University of Rochester radiology group)

Open Research Contributions

• Datasets: AVE-2 audiovisual dataset
• Benchmarks: VERIFY (COLM 2026), PW-VQA (causal VQA), MMPerspective (NeurIPS 2025)
• Code: GitHub repositories with reproducible implementations
• Demos: PromptReverb spatial audio demo

Recognition

• AAAI 2026 Best Demonstration Award Runner-up: Caption Anything in Video
• US National Interest Waiver (2023): Permanent residency for contributions to AI research
• Patent application: co-inventor, US 2025/0124292 A1 (audio-language learning)
• DARPA PTG: real-time multimodal AI development and AR demonstrations

Research focus: Egocentric and multimodal AI. Scene understanding, vision-language agents, and evaluation for real-time, on-device systems.