Max Ryabinin
mryab
AI & ML interests
Distributed training, natural language generation, efficient architectures for DL
Recent Activity
submitted
a paper
2 days ago
Untied Ulysses: Memory-Efficient Context Parallelism via Headwise Chunking published an
article
6 months ago
Fine-tune Any LLM from the Hugging Face Hub with Together AI upvoted a paper 6 months ago
When Punctuation Matters: A Large-Scale Comparison of Prompt Robustness
Methods for LLMs Organizations
