Mid-training (MT) is a training phase in LLM training pipelines that continues pretraining (PT) on smaller datasets of higher value. Taking the PT and MT phases as a whole, training usually starts with only PT data and then transitions completely to MT data. In this work, we view this complete transition from PT to MT data as a special case of a broader class of data mixing schedules that specify how the proportions of PT and MT tokens in each batch change over training. Under fixed total training compute and MT token budgets, we ask: Is this complete transition to MT data optimal for minimizing final MT loss? We show that the answer depends on the learning rate (LR) schedule: one data schedule may outperform another under one LR schedule, but changing only the LR schedule can reverse the outcome. Specifically, when the LR decays throughout continued training, introducing MT data earlier and mixing it with PT data lowers the final MT loss compared with placing all MT data at the end of training, but this result is reversed if we use constant LR. We therefore study how to jointly choose when to introduce MT data and when to decay the LR through ablation studies, and find that it is better to decay the LR only after MT data enters training. Finally, we develop a theoretical model that explains why the optimal data schedule for a fixed MT token budget depends on the LR schedule.
Xinyu Shi, Kairong Luo Tsinghua University, Zhen Zheng ByteDance Inc., Wenguang Chen Tsinghua University, Tsinghua University
Large Language Models (LLMs) have rapidly advanced in recent years, scaling up in both parameter count and context length. However, as context windows extend from thousands to hundreds of thousands of tokens, attention computation becomes the dominant source of memory usage and runtime in decoding stages, severely limiting the efficiency and scalability of long-context LLMs. Sparse attention has emerged as a promising solution, reducing complexity by computing attention over only a subset of context tokens. However, the sparse attention for Multi-head Latent Attention(MLA) which is a variant of standard MHA is rarely studied. In this paper, we introduce RoPE-based Blockwise Sparse Attention (RoBSA), a method designed specifically for MLA during the decoding stage of model inference. RoBSA leverages the decoupled nature of RoPE within MLA to implement token selection in a blockwise manner. RoBSA is a lightweight, training-free, and layer-aware algorithm that can be integrated in a plug-and-play fashion. Our method significantly reduces end-to-end inference latency in the decoding stage by up to 2.55x with minimal accuracy loss compared to full attention in long-context scenarios for very large models.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表