AAAI 2026technical0 citations
Remember Me: Bridging the Long-Range Gap in LVLMs with Three-Step Inference-Only Decay Resilience Strategies
Peng Gao, Yujian Lee, Xiaofeng Zhang, Zailong Chen, Hui Zhang
Abstract
Large Vision-Language Models (LVLMs) have achieved impressive performance across a wide range of multimodal tasks. However, they still face critical challenges in modeling long-range dependencies under the usage of Rotary Positional Encoding (ROPE). Although it can facilitate precise modeling of token positions, it induces progressive attention decay as token distance increases, especially with progressive attention decay over distant token pairs, which severely impairs the model
BibTeX
@inproceedings{aaai2026_remembermebridgi,
title = {Remember Me: Bridging the Long-Range Gap in LVLMs with Three-Step Inference-Only Decay Resilience Strategies},
author = {Peng Gao and Yujian Lee and Xiaofeng Zhang and Zailong Chen and Hui Zhang},
booktitle = {AAAI 2026},
year = {2026}
}