arXiv Machine Learning
Smaller Models, Better Rejects: Preference Distillation Scaling
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv AI
Aug 17
Capacity-Dependent Effects of Data Selection for Reasoning
arXiv:2608. 13721v1 Announce Type: cross Abstract: In reasoning supervised fine-tuning, candidate responses for the same instruction can differ substantially in how well they match the student's current distribution.
arXiv Machine Learning
Jul 30
Weak-to-Strong On-Policy Distillation
arXiv:2607. 26246v1 Announce Type: new Abstract: On-policy distillation (OPD), which aligns a student with the teacher's token-level distribution on the student's own rollouts, is an effective paradigm for transferring capabilities across LLMs.
Hugging Face Trending Papers
Jul 6
Weak-to-Strong Generalization via Direct On-Policy Distillation
Reinforcement learning with verifiable rewards (RLVR) is a powerful recipe for improving language-model reasoning, but it is expensive to repeat on every new strong model because the target model must generate many rollouts during training. As models scale, post-training itself becomes a bottleneck.
arXiv AI
Aug 11
Matching Supervision to the Student's Learning Capacity: A Unified Framework for On-Policy Self-Distillation
arXiv:2608. 08176v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) improves the reasoning abilities of LLMs by internalizing privileged context into model parameters through self-distillation.
arXiv AI
Sep 7
What Matters in On-Policy Distillation? A Perspective on Data Efficiency and Data Selection
The paper investigates data efficiency and selection in On‑Policy Distillation (OPD) for large language models. It shows that 1‑shot OPD—training on a single example—consistently improves performance, especially when the example is hard, and that longer chain‑of‑thought (CoT) paths drive the gains rather than token entropy. Based on these findings, the authors propose a simple hard‑example selection strategy that, using only eight carefully chosen hard examples, matches the performance of a 17,000‑example baseline across models from 1.5B to 7B parameters.
arXiv AI
Aug 24
Preference-Based Self-Distillation: Beyond KL Matching via Reward Regularization
The paper introduces Preference‑Based Self‑Distillation (PBSD), a new on‑policy self‑distillation method that replaces traditional KL matching with a reward‑regularized objective. PBSD derives a reward‑reweighted teacher distribution, optimizing preference gaps between teacher and student samples while keeping on‑policy sampling. Experiments on mathematical reasoning and tool‑use tasks show PBSD achieves stronger average performance, improved training stability, and maintains token efficiency compared to prior self‑distillation baselines.