🔄 In a Training Loop

Mukul Ranjan

mukul54

·

https://www.mukul54.github.io

AI & ML interests

Efficiency, Optimization

Recent Activity

upvoted a collection 2 days ago

WTF GENIUS PAPERS

authored a paper 2 days ago

Flash-dLLM: IO-Aware KV Caching and Parallel Decoding for Fast, Memory-Efficient Diffusion LLMs

upvoted a paper 2 days ago

Towards Full Pipeline FP8 Reinforcement Learning for LLMs

View all activity

Organizations

mukul54 's papers 5

arxiv:2609.26796

arxiv:2510.14973

arxiv:2505.24867

arxiv:2503.20786

arxiv:2502.14949