Scholar
Wayner Barrios
Google Scholar ID: MPJZRmsAAAAJ
Carnegie Mellon University
Citations & Impact
Citations
202
H-index
7
i10-index
6
Publications
14
Co-authors
7
list available
Publications
Resume (English only)
Academic Achievements
- Selected Publications:
- 1. MoDA: Modulation Adapter for Fine-Grained Visual Grounding in Instructional MLLMs
- 2. FT2TF: First-Person Statement Text-To-Talking Face Generation
- 3. Multi-layer Learnable Attention Mask for Multimodal Tasks
- 4. Localizing Moments in Long Video Via Multimodal Guidance
Research Experience
- Winter 2025, Research Intern at Samsung Research America, working on Reinforcement Learning and Multimodal Large Language Models; Supporting US and Latin American clients on data and business analytics (KPI decisions, market forecasting) and automation of AI/ML pipelines; Developed systems that process large-scale, real-time data for many concurrent users, leveraging distributed training and inference.
Education
- PhD Candidate, Computer Science, Dartmouth College, Advisor: SouYoung Jin
Background
- Research Interests: Multimodal Large Language Models (MLLMs), Computer Vision, Evaluation and Step-Verified Reasoning, Improved Multimodal Fusion through Learnable Masks, Cross-Modal Alignment, and Guidance Mechanisms for Long-Video Understanding. Professional Field: Video Understanding, Building End-to-End AI Datasets.
Miscellany
- Python open-source contributor, contributed to projects advancing the computer vision and geospatial communities.
