arXivLLM
Towards Looped Models Done Right: Rethinking at Fixed Points for Efficient Training, Decoding, and RL
循環語言模型的「定點」重塑：邁向高效訓練、解碼與強化學習的全新架構
This paper optimizes looped language models near their fixed points using learned depth priors and orthogonal input injection, achieving up to 1.79x faster prefill, 2x faster RL training, and 3x smaller KV cache.
2 min read