3 papers
cs.LG2026
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…
cs.AI2026
Faster IndexTTS-2: Accelerating and Streaming Autoregressive Zero-Shot Text-to-Speech Synthesis on GPUs
Muyang Du, Shuang Yu, Junjie Lai
Autoregressive text-to-speech models achieve strong naturalness but suffer from slow inference due to sequential token generation, limiting their deployment in production applicati…
cs.LG2026
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…