Blog posts

Blogs in English:

  1. AgentRunbook-C V2: A Self-Improving Lightweight Memory Controller.
  2. Reading list: long-term memory approaches.
  3. VisRet: Visualization Improves Knowledge-Intensive Text-to-Image Retrieval.
  4. Reading list: synthetic data of LLMs, by LLMs, and for LLMs.
  5. Repoformer: Selective Retrieval for Repository-Level Code Completion.
  6. Reading list: keyphrase extraction and keyphrase generation.

Blogs in Chinese:

  1. When to retrieve?
  2. Faithfulness evaluation for text summarization.
  3. Sampling strategies for NLG and their implementations.
  4. Adapters.
  5. Sequence labeling with HMM, MEMM, and CRF.

Some blogs I found interesting:

  1. Scaling Laws, Carefully
  2. An intro to einsum.
  3. The Essense of Global Convolution Models.
  4. Transformer Inference Arithmetic
  5. So Many RLHF Algorithms
  6. Diffusion Without Tears
  7. How to Scale Your Model