Cross-Tokenizer On-Policy Distillation via Byte-Prefix Marginalization
Open-weight language models from different families exhibit complementary capabilities, motivating their consolidation into a compact student through on-policy distillation (OPD). However, full-vocabulary OPD typically assumes a shared tokenizer, while existing cross-tokenizer methods may discard teacher probability mass or assign it to student tokens with unrelated content. We introduce Byte-Prefix Marginalization (BPM), which re-expresses the teacher's next-token distribution over the student
Record details
Published: 24 July 2026
Source: arXiv
Category: Research
Topics: Regulation · Children & education
Retrieved: 27 July 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
Visual Contrastive Self-Distillation
arXiv cs.AI · 23 July 2026
X$^3$-OPD: Distilling Reasoning into Large Audio-Language Models via On-Policy Alignment
arXiv cs.LG · 23 July 2026
Open Veins of Algorithmic Auditing: Why AI Assessment Lags Behind Its Deployment in the Global South
arXiv · 23 July 2026
Exploring the Design Space of LLM-Based Programming Support in CS Education: A Scoping Review through the Lens of Assistance Governance
arXiv cs.HC · 23 July 2026
Did Alice Do Wrong? Cross-Cultural Differences in Student Perceptions of Generative AI Use in University Computing Education
arXiv cs.CY · 23 July 2026
Visual Contrastive Self-Distillation
HuggingFace Daily Papers · 22 July 2026
How to cite this record
ethics.ai (24 July 2026), “Cross-Tokenizer On-Policy Distillation via Byte-Prefix Marginalization,” evidence record 13616, https://ethics.ai/record/13616 (originally published by arXiv).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.