arXiv:2606. 19004v1 Announce Type: cross Abstract: Reinforcement learning (RL) post-training of Diffusion Transformers (DiTs) is prohibitively expensive, requiring thousands of high-end GPUs.
By Ruiqi Lai, Dakai An, Wei Gao, Ju Huang, Siran Yang, Jiamang Wang, Lin Qu, Dmitrii Ustiugov, Wei Wang
arXiv:2608. 11152v1 Announce Type: cross Abstract: Modern reinforcement learning (RL) post-training pipelines for large language models (LLMs) increasingly combine rollout workloads across multiple domains and feedback paradigms.
By Zetao Hong, Song Yuan, Yuanhao Ding, Yibo Zhu, Daxin Jiang, Zhibin Wang, Chen Tian
Modern reinforcement learning (RL) post-training pipelines for large language models (LLMs) increasingly combine rollout workloads across multiple domains and feedback paradigms. Prefix-aware routing improves inference efficiency through cache reuse and load balancing, but it does not control how heterogeneous rollout sessions compete for KV-cache capacity.
arXiv:2607. 07508v1 Announce Type: cross Abstract: Reinforcement learning (RL) is becoming increasingly important for post-training large language models (LLMs).
By Zhenyu Hou, Yujiang Li, Jie Tang, Yuxiao Dong
arXiv:2607. 26253v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) is bottlenecked by rollout generation, yet many sampled prompts produce saturated groups (all responses correct or all incorrect) whose zero reward variance yields no policy-gradient signal.
By Pixel Nomand, Elena Voss, Marcus Hale, Sofia Reyes
Reinforcement learning (RL) is becoming increasingly important for post-training large language models (LLMs). Previous RL pipelines for LLMs were mostly synchronous and batch-interleaved, which is inefficient for long-horizon agentic tasks.