Large language models suffer issues when operated on long contexts that are larger than their training context length due to the standard position encoding for tokens in the attention layer. Tokens a long distance apart will rarely have an effect on each other and long prompts yield unexpected results. To solve this problem, we propose SELF (Self-Extend the Context Length With Logistic Growth Function): a solution of grouping consecutive tokens at varying group sizes using a logistic capacity equation combined with a constant group size at smaller relative distances. Our model had an increase in performance of up to 12% compared to the LongLM extension method in LEval (specifically on the Qwen model). On summarization related tasks in LongBench, our model performed up to 6.4% better than LongLM (specifically on the Llama-2-7b model). On reading comprehension tasks from LEval, our model performed up to 5.4% better than the LongLM. Our code is available atthis https URL.
View on arXiv@article{dang2025_2505.17296,
title={ SELF: Self-Extend the Context Length With Logistic Growth Function },
author={ Phat Thanh Dang and Saahil Thoppay and Wang Yang and Qifan Wang and Vipin Chaudhary and Xiaotian Han },
journal={arXiv preprint arXiv:2505.17296},
year={ 2025 }
}