Search papers, labs, and topics across Lattice.
This paper systematically investigates on-policy distillation (OPD) in large language models, revealing its role as an exploration catalyst that enhances reasoning through dense token-level guidance. The authors identify critical pathologies鈥擲tudent-Teacher Mismatch and Length Exploitation鈥攖hat hinder effective exploration and propose lightweight signal regulations to mitigate these issues. Through experiments on seven benchmarks, they demonstrate that maintaining high-quality guiding signals significantly improves OPD performance, outperforming existing variants and reinforcement learning from human feedback (RLVR) baselines.
The effectiveness of on-policy distillation hinges not on model size, but on the quality of the guiding signal, revealing hidden pitfalls that can derail exploration.
On-policy distillation (OPD) has become a key paradigm in LLM post-training, yet its training dynamics remain poorly understood. We present a systematic study examining the role, pathologies, and regulations of OPD. We first clarify the role of OPD as an exploration catalyst: it steers the student toward correct reasoning paths via dense token-level guidance, without expanding capability ceiling. We confirm this by showing that prompt diversity matters more than per-problem sampling numbers, and critically, that the effectiveness of OPD hinges entirely on the quality of its guiding signal. This dependency exposes two pathologies that derail exploration. The Student-Teacher Mismatch occurs when a large teacher-student distributional gap causes the guiding signal to misalign with task correctness, steering exploration in counterproductive directions. Length Exploitation arises when the aggregated token-level objective creates length-dependent shortcuts, allowing the student to game the reward landscape through response truncation or redundant padding, exploring degenerate length modes rather than reasoning strategies. To tame these pathologies, we investigate lightweight signal regulations: advantage clipping and log-scale compression, ensuring exploration is guided by faithful signals. Experiments across seven benchmarks demonstrate that these regulations alleviate length exploitation and enable effective distillation, stably surpassing OPD variants and RLVR baselines, thereby confirming that well-regulated signal quality, rather than mere teacher scale, governs successful exploration in OPD.
