تقرير
Generation Constraint Scaling Can Mitigate Hallucination
العنوان: | Generation Constraint Scaling Can Mitigate Hallucination |
---|---|
المؤلفون: | Kollias, Georgios, Das, Payel, Chaudhury, Subhajit |
سنة النشر: | 2024 |
المجموعة: | Computer Science |
مصطلحات موضوعية: | Computer Science - Computation and Language, Computer Science - Artificial Intelligence, Computer Science - Machine Learning |
الوصف: | Addressing the issue of hallucinations in large language models (LLMs) is a critical challenge. As the cognitive mechanisms of hallucination have been related to memory, here we explore hallucination for LLM that is enabled with explicit memory mechanisms. We empirically demonstrate that by simply scaling the readout vector that constrains generation in a memory-augmented LLM decoder, hallucination mitigation can be achieved in a training-free manner. Our method is geometry-inspired and outperforms a state-of-the-art LLM editing method on the task of generation of Wikipedia-like biography entries both in terms of generation quality and runtime complexity. Comment: 7 pages; accepted at ICML 2024 Workshop on Large Language Models and Cognition |
نوع الوثيقة: | Working Paper |
URL الوصول: | http://arxiv.org/abs/2407.16908 |
رقم الأكسشن: | edsarx.2407.16908 |
قاعدة البيانات: | arXiv |
كن أول من يترك تعليقا!