Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation
While recent work in Reinforcement Learning with Verifiable Rewards (RLVR) has shown that a small subset of critical tokens disproportionately drives reasoning gains, an analogous token-level understanding of On-Policy Distillation (OPD) remains largely unexplored. In this work, we investigate high-loss tokens, a token type that--as the most direct signal of student-teacher mismatch under OPD's per-token KL objective--should progressively diminish as training converges according to existing stud
Record details
Published: 10 May 2026
Source: arXiv
Category: Research
Topics: Regulation · Children & education · Finance, VC & PE
Retrieved: 14 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.
Knowing the Rules Is Not Enough: Student Regulatory Awareness and Use of GenAI in Higher Education
arXiv · 16 May 2026
A Systematic AI Adoption Framework for Higher Education: From Student GenAI Usage to Institutional Integration
arXiv · 23 April 2026
Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe
arXiv · 14 April 2026
Enabling and Inhibitory Pathways of Students' AI Use Concealment Intention in Higher Education: Evidence from SEM and fsQCA
arXiv · 13 April 2026
How unique are hallucinated citations offered by generative Artificial Intelligence models?
arXiv · 31 March 2026
Interpretable Markov-Based Spatiotemporal Risk Surfaces for Missing-Child Search Planning with Reinforcement Learning and LLM-Based Quality Assurance
arXiv · 9 March 2026
How to cite this record
ethics.ai (10 May 2026), “Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation,” evidence record 4640, https://ethics.ai/record/4640 (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.