GLM-5.2
Z.ai (Zhipu AI)Our RL algorithm builds upon GRPO [40] and incorporates the IcePop technique [61] to mitigate the training-inference mismatch.
implementation detail · filed under post-training
IcePop is a reinforcement-learning technique that mitigates the training-inference mismatch between the policy used for sampling trajectories and the policy used for gradient updates.1 It works by suppressing samples whose training-inference mismatch ratio deviates excessively.2
Summarised by glm-5.3-flash from 1 source; every sentence is checked against the quote it cites.
incorporates the IcePop technique [61] to mitigate the training-inference mismatch, i.e., the discrepancy between the inference distribution and the training distribution during RL optimization— GLM-5: from Vibe Coding to Agentic Engineering
The operator suppresses samples whose mismatch ratio deviates excessively— GLM-5: from Vibe Coding to Agentic Engineering
Also called IcePop technique, token-level masking using IcePop strategy.
3 quotes from 2 sources, one card per model, each opening with its strongest.
Our RL algorithm builds upon GRPO [40] and incorporates the IcePop technique [61] to mitigate the training-inference mismatch.
where represents token-level masking using IcePop strategy (ring1t)
icePop strategy (ring1t)