- Adversarial Robustness vs Continual Learning: A Gradient Conflict
- Dreaming in Latent Space: Deriving the Sequence-Level ELBO for World Models
- REFRAG: Recursive Fragmentation for Efficient Retrieval-Augmented Decoding
- Self-Supervised Learning (SSL) Deep Dive
- Mixture of Experts (MoE) [Theory and Implementation]
- KV Caching
- Prefix Tuning
- LoRA and QLoRA:Â Efficient Fine-Tuning for Large Language Models
- Towards Autonomous Preference Formation in AI: When Does “Changing One’s Mind” Become Meaningful?
- Pairwise Ranking Problem: ML Interview