arXiv AI
Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Jun 2
OPD+: Rethinking the Advantage Design for On-Policy Distillation
arXiv:2606. 01039v1 Announce Type: cross Abstract: On-policy distillation (OPD) is a widely used technique to transfer capabilities from capable teacher language models to the base student models, and can be formulated in a reinforcement learning style objective using student generated rollouts.
arXiv Machine Learning
Jul 17
On-Policy Delta Distillation
arXiv:2607. 15161v1 Announce Type: new Abstract: On-policy distillation is an alternative post-training method in reinforcement learning that alleviates the constraints imposed by reward models by providing token-level supervision from a teacher model.
arXiv Machine Learning
Jun 19
A Survey of On-Policy Distillation for Large Language Models
arXiv:2604. 00626v4 Announce Type: replace Abstract: As Large Language Models continue to grow in both capability and cost, transferring frontier capabilities into smaller, deployable students has become an important engineering problem, and knowledge distillation remains a common technique for this transfer.
Hugging Face Trending Papers
Sep 15
Beyond Token-Local Imitation: Reward-Compatible Temporal Credit Assignment for On-Policy Distillation
On-policy distillation (OPD) has emerged as an effective approach for large language model post-training, yet existing objectives face a trade-off between objective fidelity and optimization stability...
arXiv AI
Sep 16
Beyond Token-Local Imitation: Reward-Compatible Temporal Credit Assignment for On-Policy Distillation
arXiv:2609.16937v1 Announce Type: cross Abstract: On-policy distillation (OPD) has emerged as an effective approach for large language model post-training, yet existing objectives face a trade-off be...
arXiv AI
Jul 17
Beyond Entropy: Correctness-Aware Advantage Shaping via Contrastive Policy Optimization
arXiv:2607. 14614v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) commonly uses entropy for advantage shaping.