Saved in:
Bibliographic Details
Main Authors: Zhu, Tongyao, Liu, Qian, Wang, Haonan, Chen, Shiqi, Gu, Xiangming, Pang, Tianyu, Kan, Min-Yen
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.15450
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Recent advancements in LLM pretraining have featured ever-expanding context windows to process longer sequences. However, our pilot study reveals that models pretrained with shorter context windows consistently outperform their long-context counterparts under a fixed token budget. This finding motivates us to explore an optimal context window scheduling strategy to better balance long-context capability with pretraining efficiency. To this end, we propose SkyLadder, a simple yet effective approach that implements a short-to-long context window transition. SkyLadder preserves strong standard benchmark performance, while matching or exceeding baseline results on long context tasks. Through extensive experiments, we pre-train 1B-parameter models (up to 32K context) and 3B-parameter models (8K context) on 100B tokens, demonstrating that SkyLadder yields consistent gains of up to 3.7% on common benchmarks, while achieving up to 22% faster training speeds compared to baselines. The code is at https://github.com/sail-sg/SkyLadder.