Not Every Divergence Should Be Suppressed: Counterfactual Recoverability in On-Policy Distillation
cs.LG, cs.AI
Submitted: 2026-08-05
Updated: 2026-09-28
Terminology
Sources
- Distilling the Knowledge in a Neural Network
- On-Policy Distillation with Curriculum Turn-level Guidance for Multi-turn Agents
- Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe
- Policy Distillation
- SOD: Step-wise On-policy Distillation for Small Language Model Agents
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks