Ravid Shwartz-Ziv
Ravid Shwartz-Ziv
Work
Research
Writing & Podcast
Experience
Contact
Light
Dark
Automatic
Yann LeCun
Latest
HP-JEPA: Hierarchical Partitioning for Multi-Resolution Graph Joint-Embedding Predictive Learning
S-JEPA: Soft Clustering Anchors for Self-Supervised Speech Representation Learning
On Training in Imagination
AI Must Embrace Specialization via Superhuman Adaptable Intelligence
Soft Clustering Anchors for Self-Supervised Speech Representation Learning in Joint Embedding Prediction Architectures
JEPA as a Neural Tokenizer: Learning Robust Speech Representations with Density Adaptive Attention
Attention Sinks and Compression Valleys in LLMs are Two Sides of the Same Coin
Layer by Layer: Uncovering Hidden Representations in Language Models
From Tokens to Thoughts: How LLMs and Humans Trade Compression for Meaning
Rate-In: Information-Driven Adaptive Dropout Rates for Improved Inference-Time Uncertainty Estimation
Seq-VCR: Preventing Collapse in Intermediate Transformer Representations for Enhanced Reasoning
LiveBench: A Challenging, Contamination-Limited LLM Benchmark
To Compress or Not to Compress--Self-Supervised Learning and Information Theory: A Review
What Do We Maximize in Self-Supervised Learning?
Pre-Train Your Loss: Easy Bayesian Transfer Learning with Informative Priors
Cite
×