一句话：提出 GlitchProber，面向大语言模型词表中的 Glitch Token 进行有效检测与缓解，以提升模型可靠性。
大语言模型（LLM）在自然语言处理中取得巨大成功，但其内部机制的黑盒性引发了对可信性与可解释性的担忧。近期研究在模型词表空间中发现一类异常 token，称为 Glitch Token；一旦这些 token 出现在输入中，可能诱导模型产生错误、无关甚至有害的结果，严重削弱 LLM 的可靠性与实用性。本文提出 GlitchProber，旨在推进对 Glitch Token 的有效检测与缓解。摘要未给出具体检测方法、缓解策略及实验评估细节，但标题表明该工作聚焦于这一异常词元问题的工具化检测与修复，属于提升 LLM 可靠性的研究。
Large language models (LLMs) have achieved unprecedented success in the field of natural language processing. However, the black-box nature of their internal mechanisms has brought many concerns about their trustworthiness and interpretability. Recent research has discovered a class of abnormal tokens in the model's vocabulary space and named them "glitch tokens". Those tokens, once included in the input, may induce the model to produce incorrect, irrelevant, or even harmful results, drastically undermining the reliability and practicality of LLMs.
@inproceedings{paperbot8336,
title = {GlitchProber: Advancing Effective Detection and Mitigation of Glitch Tokens in Large Language Models},
author = {Zhibo Zhang and Wuxia Bai and Yuxi Li and Mark Huasong Meng and Kailong Wang and Ling Shi and Li Li and Jun Wang and Haoyu Wang},
booktitle = {ASE},
year = {2024},
doi = {10.1145/3691620.3695060}
}