EP281: Restoring plasticity to over-trained AI

Title: When RL Fails after SFT: Rejuvenating Model Plasticity for Robust SFT-to-RL HandoffSource: http://arxiv.org/abs/2606.09932v1Summary:This paper identifies and solves the critical 'loss of plasticity' bottleneck in the standard LLM post-training pipeline where excessive SFT inhibits subsequent RL optimization. It introduces 'Rejuvenation', a foundational training primitive that uses model fusion and neuron resets to enable robust reasoning gains during RL while preserving SFT-acquired knowledge.