ARTFEED — Contemporary Art Intelligence

Hera: Step-Level Device-Cloud Coordination for LLM Agents

ai-technology · 2026-05-26

Researchers propose Hera, a step-level coordinator for device-cloud LLM agents tackling long-horizon tasks. It uses a two-stage training paradigm: imitation learning for cold-start, then reinforcement learning optimizing both task success and cloud usage. This achieves a strong performance-cost Pareto frontier, addressing the coarse task-level routing limitations of existing systems.

Key facts

  • Hera is a step-level device-cloud LLM agent coordinator.
  • It targets long-horizon tasks.
  • Uses two-stage training: imitation learning then reinforcement learning.
  • Optimizes task success and cloud usage efficiency.
  • Achieves strong performance-cost Pareto frontier.
  • Addresses limitations of coarse task-level routing.
  • Published on arXiv with ID 2605.24598.
  • Focuses on device-cloud dilemma for LLM agents.

Entities

Institutions

  • arXiv

Sources