Scholar
Ashutosh Chaubey
Google Scholar ID: 8g_xYb0AAAAJ
CS PhD, University of Southern California
Citations & Impact
Citations
91
H-index
4
i10-index
1
Publications
10
Co-authors
4
list available
Publications
Resume (English only)
Academic Achievements
- Paper 'Face-LLaVA: Facial Expression and Attribute Understanding through Instruction Tuning' accepted at WACV 2026 (Round-1 early acceptance, 6.4% acceptance rate)
- Paper accepted at EMNLP 2025 (Findings)
- Paper 'DiTaiListener: Controllable High Fidelity Listener Video Generation with Diffusion' accepted at ICCV 2025
- Paper accepted at WACV 2025
- Paper accepted at ASRU 2023
- Paper accepted at Interspeech 2022
Research Experience
- PhD Researcher, Intelligent Human Perception Lab, Institute for Creative Technologies, USC (since Aug 2024)
- Founding Research Engineer at Anoki AI (from Apr 2023), working on multimodal content understanding and retrieval
- Data Scientist at LG Ad Solutions (from Jul 2021), working on speaker recognition, automatic content recognition using audio, and voice cloning
- Interned at Adobe Research
- Interned at Vision and AI Lab, IISc Bengaluru, advised by Prof. R. Venkatesh Babu
- Interned at IIT Roorkee, advised by Prof. R. Balasubramanian
Background
- CS PhD student at the Institute for Creative Technologies, University of Southern California
- Advised by Prof. Mohammad Soleymani at the Intelligent Human Perception Lab
- Research focuses on post-training techniques (e.g., preference optimization) for multimodal (audio and video) LLMs to enhance social and emotion understanding
- Collaborates on diffusion-based video generation projects for modeling social behaviors
- Research interests: Multimodal LLM tuning and post-training, emotion understanding, Social AI
