PIVOT: Perplexity-Informed KD-to-RL Transition Scheduling for Vertical-Domain Few-Shot Distillation
cs.LG, cs.AI
Submitted: 2026-10-08
Updated: 2026-10-08
Terminology
Sources
- On-Policy Distillation of Language Models: Learning from Self-Generated Mistakes
- HDPO: Hybrid Distillation Policy Optimization via Privileged Self-Distillation
- Improving Generalization in Intent Detection: GRPO with Reward-Based Curriculum Sampling
- Revisiting On-Policy Distillation: Empirical Failure Modes and Simple Fixes
- Distilling the Knowledge in a Neural Network
- Entropy-Aware On-Policy Distillation of Language Models
- CoDistill-GRPO: A Co-Distillation Recipe for Efficient Group Relative Policy Optimization
- Unifying Group-Relative and Self-Distillation Policy Optimization via Sample Routing
- Benchmarking Natural Language Understanding Services for building Conversational Agents
- Teacher-Guided Policy Optimization for On-Policy Reasoning Distillation under Large Policy Divergence
- SuperRL: Reinforcement Learning with Supervision to Boost Language Model Reasoning
- DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- KDRL: Post-Training Reasoning LLMs via Unified Knowledge Distillation and Reinforcement Learning
- PACED: Distillation and On-Policy Self-Distillation at the Frontier of Student Competence
- Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation
- Multi-Rollout On-Policy Distillation via Peer Successes and Failures
- SCOPE: Signal-Calibrated On-Policy Distillation Enhancement with Dual-Path Adaptive Weighting
- DianJin-R1: Evaluating and Enhancing Financial Reasoning in Large Language Models
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: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks