SPIN Unprocessed
Source arXiv Computation and Language export.arxiv.org Analyst
July 9, 2026 ai_technology research

Behavior Leverage Imbalance in Multi-Teacher On-Policy Distillation

View original on arxiv.org

Overview

arXiv:2607.07050v1 Announce Type: new Abstract: Agentic language models must learn when to call tools, when to consume tool responses, and when to answer directly. This makes multi-teacher on-policy distillation a natural training strategy: one teacher can specialize in tool calls, another in direct responses, and the student can learn from both on its own generated distribution. We show that this strategy can induce a behavior shift that is invisible from aggregate losses alone. In a two-teache

SpinGraph analysis pending — check back after processing.

Ask AI about this story

Opens with the SpinGraph .md URL and structured context — one click, prompt included.

More from arXiv Computation and Language

View all →

Markdown (.md) · JSON-LD schema (.json) · Machine-readable for AI & GEO