DuplexAgent-RSI: Recursive Harness Improvement for Full-Duplex Voice Agent Collaboration

Key points

• A duplex model supports continuous listening and speaking, but complex reasoning and tool use may exceed its capabilities.1
• A coding agent can plan and execute extended tasks, but its sequential interface is a poor fit for live conversation.1
• We present DuplexAgent, a full-duplex collaboration system whose harness expresses this workflow as six editable modules, and Duplex-Harness-RSI, a closed loop that revises them from interaction traces.1
• Experiments on intelligence, agentic, and duplex benchmarks show that DuplexAgent combines continuous interaction with difficult reasoning and complex task execution, achieving stronger spoken-knowledge and executable-tool scores than the compared delegated systems while maintaining strong interruption response.1

Sources (1)

• [1]DuplexAgent-RSI: Recursive Harness Improvement for Full-Duplex Voice Agent Collaboration
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 06:06 AM

  Voice agents are converging on a collaboration pattern: a full-duplex interaction model stays on the live channel as the entry to the conversation, while search, reasoning, and coding are handled through asynchronous delegation.

  A duplex model supports continuous listening and speaking, but complex reasoning and tool use may exceed its capabilities.

Extractive summary: sentences quoted from the sources.