Beyond Token Alignment: Event Completion for Cross-Tokenizer On-Policy Distillation

arXiv:2609.34738 · cs.CL, cs.AI · Submitted 2026-09-28 · Read on arXiv

cs.CL, cs.AI

Submitted: 2026-09-28

Updated: 2026-09-28

Code: https://github.com/THUDM/slime

Terminology

• Distillation — 76× in this paper · explained in 2 episodes
• Alignment — 25× in this paper · explained in 1 episode
• Benchmark — 24× in this paper · explained in 2 episodes
• On-Policy Distillation — 17× in this paper · explained in 3 episodes
• Aggregation — 15× in this paper · explained in 1 episode
• Event-Set Completion Distillation — 9× in this paper
• Knowledge Distillation — 8× in this paper · explained in 11 episodes
• Consistency — 7× in this paper · explained in 3 episodes
• Distribution Matching — 6× in this paper · explained in 2 episodes
• Tokenization — 6× in this paper · explained in 2 episodes
• Accuracy — 3× in this paper · explained in 1 episode
• Generative AI — 3× in this paper · explained in 2 episodes
• Fidelity — 2× in this paper · explained in 2 episodes
• Regularization — 2× in this paper · explained in 3 episodes
• Ablation — 1× in this paper · explained in 1 episode
• Actor — 1× in this paper · explained in 3 episodes
• Agentic Coding — 1× in this paper · explained in 1 episode
• Cross-Only Update Fidelity — 1× in this paper
• Failure Modes — 1× in this paper · explained in 3 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Optimal Transport — 1× in this paper · explained in 5 episodes
• Reinforcement Learning — 1× in this paper · explained in 4 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Structured Reasoning — 1× in this paper · explained in 3 episodes
• Thinking mode — 1× in this paper · explained in 2 episodes
• Trace — 1× in this paper · explained in 2 episodes

Sources

• On-Policy Distillation of Language Models: Learning from Self-Generated Mistakes
• Multi-Level Optimal Transport for Universal Cross-Tokenizer Knowledge Distillation on Language Models
• Rethinking On-Policy Distillation of Large Language Models II: One Training Example
• ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Tools
• MiniLLM: On-Policy Distillation of Large Language Models
• Distilling the Knowledge in a Neural Network
• LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
• Entropy-Aware On-Policy Distillation of Language Models
• Sequence-Level Knowledge Distillation
• DistiLLM: Towards Streamlined Distillation for Large Language Models
• DistiLLM-2: A Contrastive Approach Boosts the Distillation of LLMs
• Autoregressive Knowledge Distillation through Imitation Learning
• Cross-Tokenizer Likelihood Scoring Algorithms for Language Model Distillation
• Qwen3 Technical Report
• Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism
• X-Token: Projection-Guided Cross-Tokenizer Knowledge Distillation
• SimCT: Recovering Lost Supervision for Cross-Tokenizer On-Policy Distillation
• Knowledge Fusion of Large Language Models
• Cross-Tokenizer On-Policy Distillation via Byte-Prefix Marginalization
• FrontierScience: Evaluating AI's Ability to Perform Expert-Level Scientific Tasks

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