On-Policy Distillation Teaches New Skills but Not New Knowledge

arXiv:2610.09639 · cs.CL · Submitted 2026-10-07 · Read on arXiv

cs.CL

Submitted: 2026-10-07

Updated: 2026-10-07

Code: https://github.com/project-numina/aimo-progress-prize

Terminology

• Distillation — 1× in this paper · explained in 2 episodes
• On-Policy Distillation — 1× in this paper · explained in 3 episodes

Sources

• Does On-Policy Distillation Really Distill? From Noisy Teacher to Self-Improvement
• Towards Understanding On-Policy Distillation through the Lens of Test-Time Scaling
• Distilling the Knowledge in a Neural Network
• Tulu 3: Pushing Frontiers in Open Language Model Post-Training
• Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe
• Olmo 3
• Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context
• Gemma 2: Improving Open Language Models at a Practical Size
• Qwen3 Technical Report
• Demystifying On-Policy Distillation: Roles, Pathologies, and Regulations

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering