Addressing Performance Saturation for LLM RL via Precise Entropy Curve Control
Abstract
Entrocraft, a rejection-sampling approach for reinforcement learning, addresses performance saturation in large language models by customizing entropy schedules and improving generalization and training longevity.
Reinforcement learning (RL) has enabled complex reasoning abilities in large language models (LLMs). However, most RL algorithms suffer from performance saturation, preventing continued gains as RL training scales. This problem can be characterized by the collapse of entropy, a key diagnostic for exploration in RL. Existing attempts focus on preventing entropy collapse through regularization or clipping. However, their resulting entropy curves often exhibit instability in the long term, which hinders performance gains. In this paper, we introduce Entrocraft, a simple rejection-sampling approach that realizes user-customized entropy schedule by biasing the advantage distributions. Entrocraft requires no objective regularization and is advantage-estimator-agnostic. Theoretically, we relate per-step entropy change to the advantage distribution under minimal assumptions. This explains the behavior of existing RL and entropy-preserving methods. Entrocraft also enables a systematic study of entropy schedules, which reveals that linear annealing, which starts high and decays to a slightly lower target, performs best. Empirically, Entrocraft addresses performance saturation, significantly improving generalization, output diversity, and long-term training. It enables a 4B model to outperform an 8B baseline, sustains improvement for up to 4x longer before plateauing, and raises pass@K by 50% over the baseline.
Community
What is the optimal entropy curve for LLM RL to improve in the long term? We introduce the best entropy curve and the trick to achieve it for any policy-gradient method.
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- Rethinking Exploration in RLVR: From Entropy Regularization to Refinement via Bidirectional Entropy Modulation (2026)
- Targeted Exploration via Unified Entropy Control for Reinforcement Learning (2026)
- SPS: Steering Probability Squeezing for Better Exploration in Reinforcement Learning for Large Language Models (2026)
- AEM: Adaptive Entropy Modulation for Multi-Turn Agentic Reinforcement Learning (2026)
- A Comparative Theoretical Analysis of Entropy Control Methods in Reinforcement Learning (2026)
- Uniform-Correct Policy Optimization: Breaking RLVR's Indifference to Diversity (2026)
- Policy Split: Incentivizing Dual-Mode Exploration in LLM Reinforcement with Dual-Mode Entropy Regularization (2026)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend
Get this paper in your agent:
hf papers read 2604.26326 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper