Mitigating the Length-Scaling Tax with Online Distillation
cs.LG, cs.AI
Submitted: 2026-09-30
Updated: 2026-09-30
Terminology
Sources
- L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning
- Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
- BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
- Small Generalizable Prompt Predictive Models Can Steer Efficient RL Post-Training of Large Reasoning Models
- Contrastive On-Policy Distillation
- CRISP: Compressed Reasoning via Iterative Self-Policy Distillation
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters
- The Shadow Price of Reasoning: Economic Perspective on Optimal Budget Allocation for LLMs
- Self-Consistency Improves Chain of Thought Reasoning in Language Models
- Just Enough Thinking: Efficient Reasoning with Adaptive Length Penalties Reinforcement Learning
- Qwen3 Technical Report
- Demystifying Long Chain-of-Thought Reasoning in LLMs
- DAPO: An Open-Source LLM Reinforcement Learning System at Scale
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