Scholar
Yingtian Tang
Google Scholar ID: 4096COYAAAAJ
EPFL
Citations & Impact
Citations
70
H-index
4
i10-index
2
Publications
8
Co-authors
0
Publications
Resume (English only)
Academic Achievements
- Diverse Perceptual Representations Across Visual Pathways Emerge from A Single Objective, bioRxiv.
- From Language to Cognition: How LLMs Outgrow the Human Language Network, accepted at CCN 2025.
- Dreaming Out Loud: A Self-Synthesis Approach For Training Vision-Language Models With Developmentally Plausible Data, BabyLM Challenge at CoNLL 2024.
- Online Motion Style Transfer for Interactive Character Control, Arxiv Preprint (2021).
- Learning-Aided Heuristics Design for Storage System, ACM SIGMOD/PODS 2021.
- Accurate probabilistic miss ratio curve approximation for adaptive cache allocation in block storage systems, DATE 2022.
- Visual Analytic System for Pandemic Management During COVID-19, winner of ICIP 2020 IEEE Signal Processing Society 5-Minute Video Clip Contest.
Background
- Currently a PhD student at EPFL, working on NeuroAI with Prof. Martin Schrimpf.
- Interested in artificial intelligence, especially how current machine learning methods can achieve active perception and learning in the real world.
- Believes this research lies at the intersection of computational neuroscience, computational cognitive science, and machine learning.
- Inspired by the work of James J. Gibson and Peter Dayan, and believes machine learning will advance their theories or be guided by them.
- Currently working on building models for dynamic perceptual representation based on state-of-the-art representation learning methods (e.g., JEPA, MAE) and brain recordings during video watching.
Co-authors