InfoPO: Information-Driven Policy Optimization for User-Centric Agents
Real-world user requests to LLM agents are often underspecified. Agents must interact to acquire missing information and make correct downstream decisions. However, current multi-turn GRPO-based methods often rely on trajectory-level reward computation, which leads to credit assignment problems and insufficient advantage signals within rollout groups. A feasible approach is to identify valuable interaction turns at a fine granularity to drive more targeted learning. To address this, we introduce InfoPO (Information-Driven Policy Optimization), which frames multi-turn interaction as a process of active uncertainty reduction and computes an information-gain reward that credits turns whose f...
| Authors | Fanqi Kong, Jiayi Zhang, Mingyi Deng et al. |
|---|---|
| Primary category | cs.AI |
| Published | 28 February 2026 |
| Last updated at source | 17 June 2026 |
Connected in the knowledge graph
- phanviethoang1512/Qwen3-8B-InfoPO-AgentModel · cited by
Models and datasets referencing this paper
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.