Unimodal Augmentation-based SSL Objectives Induce Cross-Modal Alignment
cs.LG, stat.ML
Submitted: 2026-11-02
Updated: 2026-11-02
Code: https://github.com/shaham-lab/Aug-SSL-Implicit-Alignment
Terminology
Sources
- A quantitative analysis of semantic information in deep representations of text and images
- Representation Alignment Rests on Linear Structure
- Revisiting Feature Prediction for Learning Visual Representations from Video
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- The Platonic Universe: Do Foundation Models See the Same Sky?
- Universally Converging Representations of Matter Across Scientific Foundation Models
- Revisiting the Platonic Representation Hypothesis: An Aristotelian View
- Phi-3 Safety Post-Training: Aligning Language Models with a "Break-Fix" Cycle
- The Platonic Representation Hypothesis
- Causes and Consequences of Representational Similarity in Machine Learning Models
- RoBERTa: A Robustly Optimized BERT Pretraining Approach
- Representation Learning with Contrastive Predictive Coding
- DINOv2: Learning Robust Visual Features without Supervision
- Qwen2.5 Technical Report
- Gemma 4 Technical Report
- Understanding the Emergence of Multimodal Representation Alignment
- LLaMA: Open and Efficient Foundation Language Models
- Signal-to-Noise Ratio and Sample Size Govern Representational Alignment in Neural Networks
- Escaping Plato's Cave: JAM for Aligning Independently Trained Vision and Language Models
- Proof of a perfect platonic representation hypothesis
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis