Which Tasks Survive Self-Supervised Learning?

arXiv:2609.38393 · cs.LG, cs.AI · Submitted 2026-09-29 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-09-29

Updated: 2026-09-29

Code: https://github.com/deepmind/3d-shapes

Terminology

• Self-Supervised Learning — 1× in this paper · explained in 8 episodes

Sources

• A Cookbook of Self-Supervised Learning
• Generalization Bounds for Few-Shot Transfer Learning with Pretrained Classifiers
• Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
• Unraveling Projection Heads in Contrastive Learning: Insights from Expansion and Shrinkage
• Understanding and Improving the Role of Projection Head in Self-Supervised Learning
• Directional Neural Collapse Explains Few-Shot Transfer in Self-Supervised Learning
• SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features
• Contextures: The Mechanism of Representation Learning

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks