AI RESEARCH
Hierarchical Policy Optimization for Simultaneous Translation of Unbounded Speech
arXiv CS.CL
•
ArXi:2604.21045v1 Announce Type: new Simultaneous speech translation (SST) generates translations while receiving partial speech input. Recent advances show that large language models (LLMs) can substantially improve SST quality, but at the cost of high computational overhead. To reduce this cost, prior work reformulates SST as a multi-turn dialogue task, enabling full reuse of the LLM's key-value (KV) cache and eliminating redundant feature recomputation.