TL;DR：太長了，重新加權以實現高效的大型語言模型推理壓縮

TL;DR: Too Long, Do Re-weighting for Effcient LLM Reasoning Compression

June 3, 2025

作者: Zhong-Zhi Li, Xiao Liang, Zihao Tang, Lei Ji, Peijie Wang, Haotian Xu, Xing W, Haizhen Huang, Weiwei Deng, Ying Nian Wu, Yeyun Gong, Zhijiang Guo, Xiao Liu, Fei Yin, Cheng-Lin Liu

cs.AI

摘要

大型語言模型（LLMs）近期通過強化學習和擴展的思維鏈（CoT）技術取得了顯著進展。然而，在進行高效語言推理——特別是在生成極長輸出時的推理——這一挑戰已引起研究界越來越多的關注。在本研究中，我們提出了一種基於動態比例的訓練流程，該流程不依賴於複雜的數據註釋或多個模型之間的插值。我們持續平衡模型系統1和系統2數據之間的權重，以消除冗餘的推理過程，同時保留模型的推理能力。我們在DeepSeek-R1-Distill-7B和DeepSeek-R1-Distill-14B模型上以及一系列難度各異的基準測試中驗證了我們的方法。我們的方法在保持推理準確性的同時，顯著減少了近40%的輸出標記數量。我們的代碼和數據將很快公開。

English

Large Language Models (LLMs) have recently achieved remarkable progress by leveraging Reinforcement Learning and extended Chain-of-Thought (CoT) techniques. However, the challenge of performing efficient language reasoning--especially during inference with extremely long outputs--has drawn increasing attention from the research community. In this work, we propose a dynamic ratio-based training pipeline that does not rely on sophisticated data annotations or interpolation between multiple models. We continuously balance the weights between the model's System-1 and System-2 data to eliminate redundant reasoning processes while preserving the model's reasoning capability. We validate our approach across models on DeepSeek-R1-Distill-7B and DeepSeek-R1-Distill-14B and on a diverse set of benchmarks with varying difficulty levels. Our method significantly reduces the number of output tokens by nearly 40% while maintaining the accuracy of the reasoning. Our code and data will be available soon.