From Uncertainty to Action: Learning to Steer LLM Agents
Steering an LLM agent means deciding whether to correct it, at which step, and with which mechanism.
ProofPaper ↗
Key points
- Uncertainty is often used to decide when to correct an agent, but whether it can guide these decisions remains unclear.
- We steer agent trajectories separately at every non-terminal step with each of four mechanisms and run each continuation to completion.
- We therefore propose VoS (Value of Steering), a trajectory-level monitor, offline or online, that learns from SOT the value of steering at each step and decides where to steer by it.
- A harm-budgeted trigger decides whether to steer, limiting the fraction of successful trajectories that VoS disturbs.
Sources (1)
- [1]From Uncertainty to Action: Learning to Steer LLM AgentsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 09:05 PM
Steering an LLM agent means deciding whether to correct it, at which step, and with which mechanism.
Uncertainty is often used to decide when to correct an agent, but whether it can guide these decisions remains unclear.
Extractive summary: sentences quoted from the sources.
