• Posts
Markdown Sample
Shortcodes Sample
Internship
• MiniMind
LLM
• LLM architecture
  • Encoder-only
Paper Reading
• dLLM
• YAN
Pytorch
Pytorch
Transformer
• Tokenizer
• Word Embedding
• Model
  • LSTM
  • RNN
algorithm
• Array
  • Binary Search
• Backtracking
• Binary Tree
• Dyanmic Programming
  • Multi-dimensional Dynamic Programming
• Graph Theory
• Linked List
• Two-pointers
  • Leetcode 27. Remove Element
  • Leetcode 541. Reverse String II
  • Leetcode 344. Reverse String
  • Sliding Window

Diffusion LLM paper reading

1 Residual Context Diffusion Language Models Method Traditional dLLMs discard the [MASK] at unpredicted positions in each round and reset them as new blank [MASK]s. To avoid wasting computational power, this method retains these discarded semantics and transforms them into vectors for the next round of prediction. Algorithm Generate the residual vector by performing a weighted sum of the model’s predicted probability and the word embeddings: $\Delta_i^{(t_k)} = \sum_{j=1}^V p_{i,j}^{(t_k)} E_{j,:}$

Saturday, March 14, 2026 | 8 minutes Read

Yan:Foundational Interactive Video Generation

Abstract Introduction The first paragraph primarily discusses the applications and deficiencies of Interactive Generative Video (IGV). AIGC is evolving from generating text and images to video synthesis, and has now progressed to IGV. IGV requires dynamic reactions to user input, with applications ranging from virtual simulation to embodied intelligence. However, the deficiencies of current methods lie in the lack of high visual fidelity, sustained temporal coherence, and rich interactivity. Generated content also remains static after creation, failing to adapt in real-time. (Not high-definition, not coherent, not interactive).

Monday, February 9, 2026 | 12 minutes Read