SOURCE-LINKED INTELLIGENCE

When EOS Tokens Disagree: Understanding Length Inflation in On-Policy Distillation

arXiv · AI, language, vision and robotics · article · Sep 17, 2026 · UTC

We study length inflation in on-policy distillation (OPD), where student responses can become excessively long and even exhaust the generation budget. We identify \emph{termination-token mismatch} between base students and post-trained teachers as an important source of this behavior. Across Qwen3, Llama, and Gemma, the two models can place their stopping probability on different EOS tokens, even when their declared stopping sets are identical. This mismatch can suppress the student's preferred termination action without reliably transferring the teacher-preferred alternative. We show that ali

Read original source ↗ Open in workspace

recordType

paper

region

Global