6 papers
TRIAGE: Direction-Aware Mismatch Stabilization of Native NVFP4 Reinforcement Learning
Zhen Li, Shuai Zhang, Yanggan Gu +7
Low-precision execution can substantially accelerate reinforcement learning (RL) for large language models, but discrepancies between learner and sampler execution can destabilize…
Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training
Zili Wang, Zhaopeng Qiu, Yuekai Zhang +2
Speculative decoding accelerates rollout generation, which dominates the cost of reinforcement learning (RL) post-training. Online co-training can further increase the draft's accu…
A GPU-Parallel Framework for Heterogeneous Multi-Task Reinforcement Learning
Rui Zhang, Qiwei Wu, Zhengyu Zhang +7
GPU-parallel simulation provides abundant robot interaction, but existing benchmarks rarely combine this scale with heterogeneous manipulation tasks and standardized multi-task RL…
AI-Model Network: Concept, Current State and Future
Li Zhetao, Zeng Xiyu, Wang Jianhui +6
While the primary function of computers lies in computation and processing, the core value of the Internet is rooted in sharing and collaboration. Computers create the Internet, an…
AIConfigurator: Lightning-Fast Configuration Optimization for Multi-Framework LLM Serving
Tianhao Xu, Yiming Liu, Xianglong Lu +18
Optimizing Large Language Model (LLM) inference in production systems is increasingly difficult due to dynamic workloads, stringent latency/throughput targets, and a rapidly expand…
FP8-RL: A Practical and Stable Low-Precision Stack for LLM Reinforcement Learning
Zhaopeng Qiu, Shuang Yu, Jingqi Zhang +4
Reinforcement learning (RL) for large language models (LLMs) is increasingly bottlenecked by rollout (generation), where long output sequence lengths make attention and KV-cache me…