Entropy Interventions Do Not Transfer: Why RLVR Entropy Control Misfires in On-Policy Distillation
Open Access
- 6 November 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 7503-7508
Abstract
No abstract availableThis publication has 18 references indexed in Scilit:
- HumanLLM: Towards Personalized Understanding and Simulation of Human Nature2026
- SpecQuant: Spectral Decomposition and Adaptive Truncation for Ultra-Low-Bit LLMs QuantizationProceedings of the AAAI Conference on Artificial Intelligence, 2026
- FROG: Fair Removal on Graph2025
- Towards Expansive and Adaptive Hard Negative Mining: Graph Contrastive Learning via Subspace Preserving2024
- Multi-Class Support Vector Machine with Maximizing Minimum MarginProceedings of the AAAI Conference on Artificial Intelligence, 2024
- Back to Basics: Revisiting REINFORCE-Style Optimization for Learning from Human Feedback in LLMs2024
- Joint Admission Control and Resource Allocation of Virtual Network Embedding via Hierarchical Deep Reinforcement LearningIEEE Transactions on Services Computing, 2023
- f-Divergence Minimization for Sequence-Level Knowledge Distillation2023
- Explain the Explainer: Interpreting Model-Agnostic Counterfactual Explanations of a Deep Reinforcement Learning AgentIEEE Transactions on Artificial Intelligence, 2022
- R e LAX: Reinforcement Learning Agent Explainer for Arbitrary Predictive Models2022