TensorHub: Scalable and Elastic Weight Transfer for LLM RL Training
Authors:
Open Access
- 28 September 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 882-897
Abstract
No abstract availableAll Related Versions
This publication has 12 references indexed in Scilit:
- Laminar: A Scalable Asynchronous RL Post-Training Framework2026
- DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learningNature, 2025
- HybridFlow: A Flexible and Efficient RLHF Framework2025
- RL-GPT: Integrating Reinforcement Learning and Code-as-policy2024
- GEMINI: Fast Failure Recovery in Distributed Training with In-Memory Checkpoints2023
- Efficient Memory Management for Large Language Model Serving with PagedAttention2023
- Hoplite2021
- FoundationDB: A Distributed Unbundled Transactional Key Value StoreACM SIGMOD Record, 2021
- UCX: An Open Source Framework for HPC Network APIs and Beyond2015
- Serverless network file systems1995