arXiv:2611.01987 · v1

Same Trajectory, Different Knowledge: Information-State-Aware On-Policy Distillation for Search Agents

首次提交 2026-11-01 23:24

版本更新 2026-11-01 23:24

雷达观测 2026-11-04 23:43

摘要

On-policy distillation trains search agents with teacher feedback on student-generated trajectories. Recent methods augment the teacher's context with supporting evidence, reference answers, or hindsight information. However, this privileged information may allow the teacher to rely on facts that the student cannot reliably recover from its current state. The resulting information-state mismatch can lead the teacher to undervalue retrieval that helps the student establish those facts. To address this mismatch, we introduce Information-State-Aware Distillation (ISA-Distill), which adapts teacher feedback to the student's factual recoverability. ISA-Distill estimates how reliably the student can recover action-relevant knowledge and derives teacher feedback for the same sampled action from two complementary views. The fact-use view supplies the teacher with the relevant fact, producing feedback that guides the student in using knowledge. The evidence-seeking view instead presents the information need, producing feedback that guides the student in acquiring supporting evidence. The two signals are weighted toward fact use when recoverability is high and toward evidence seeking when it is low, and the resulting feedback guides policy optimization. Experiments across diverse QA benchmarks and model scales demonstrate improved answer accuracy and retrieval efficiency. Our code and model checkpoints will be open-sourced.