Jason Ramapuram
11 papers · 2020–2025 · 4 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓+4 more ↓ Show less ↑
π§ Keyword Pioneer πΊοΈ Taxonomy Completionist (13) π£ Hot Topic Early Bird π Conference Polyglot (4) π Academic Marathon (5)
π
Cross-Pollinator
(12)
π
Interdisciplinary Bridge
π
Century Club
(11)
π₯
Unstoppable
(6)
Conferences
ICML (5)
ICLR (3)
NIPS (2)
ICCV (1)
Top co-authors
Keywords
self-supervised learning
(5)
representation learning
(3)
contrastive learning
(2)
embedding learning
(1)
multimodal learning
(1)
audio-visual learning
(1)
image reconstruction
(1)
neural network optimization
(1)
mutual information
(1)
multi-view learning
(1)
multiview learning
(1)
learning rate
(1)
training stability
(1)
masked autoencoder
(1)
equivariant representation
(1)
position embedding
(1)
layer normalization
(1)
synthetic dataset
(1)
exponential moving average
(1)
video representation
(1)
Papers
Distillation Scaling Laws
ICML 2025
Theory, Analysis, and Best Practices for Sigmoid Self-Attention
ICLR 2025
Poly-View Contrastive Learning
ICLR 2024
The Role of Entropy and Reconstruction in Multi-View Self-Supervised Learning
ICML 2023
How to Scale Your EMA
NIPS 2023
Stabilizing Transformer Training by Preventing Attention Entropy Collapse
ICML 2023
DUET: 2D Structured and Approximately Equivariant Representations
ICML 2023
Position Prediction as an Effective Pretraining Strategy
ICML 2022
Hypersim: A Photorealistic Synthetic Dataset for Holistic Indoor Scene Understanding
ICCV 2021
Kanerva++: Extending the Kanerva Machine With Differentiable, Locally Block Allocated Latent Memory
ICLR 2021
Self-Supervised MultiModal Versatile Networks
NIPS 2020