English

LIFT: Improving Long Context Understanding Through Long Input Fine-Tuning

Computation and Language 2024-12-19 v1 Artificial Intelligence

Abstract

Long context understanding remains challenging for large language models due to their limited context windows. This paper introduces Long Input Fine-Tuning (LIFT) for long context modeling, a novel framework that enhances LLM performance on long-context tasks by adapting model parameters to the context at test time. LIFT enables efficient processing of lengthy inputs without the computational burden of offline long-context adaptation, and can improve the long-context capabilities of arbitrary short-context models. The framework is further enhanced by integrating in-context learning and pre-LIFT supervised fine-tuning. The combination of in-context learning and LIFT enables short-context models like Llama 3 to handle arbitrarily long contexts and consistently improves their performance on popular long-context benchmarks like LooGLE and LongBench. We also provide a comprehensive analysis of the strengths and limitations of LIFT on long context understanding, offering valuable directions for future research.

Keywords

Cite

@article{arxiv.2412.13626,
  title  = {LIFT: Improving Long Context Understanding Through Long Input Fine-Tuning},
  author = {Yansheng Mao and Jiaqi Li and Fanxu Meng and Jing Xiong and Zilong Zheng and Muhan Zhang},
  journal= {arXiv preprint arXiv:2412.13626},
  year   = {2024}
}
R2 v1 2026-06-28T20:40:06.807Z