Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding
Paper • 2604.26779 • Published • 13
Hayate Iso
isomap
·
AI & ML interests
Large Language Models, Distributed Systems, Text Generation
Recent Activity
authored a paper 16 days ago
Retrieval Helps or Hurts? A Deeper Dive into the Efficacy of Retrieval
Augmentation to Language Models authored a paper 16 days ago
Holistic Reasoning with Long-Context LMs: A Benchmark for Database
Operations on Massive Textual Data authored a paper 16 days ago
Comparative Opinion Summarization via Collaborative DecodingOrganizations
upvoted a paper 17 days ago
upvoted an article 6 months ago
Article
PipelineRL
ServiceNow
• • 44 upvoted a paper almost 3 years ago