Generation Constraint Scaling Can Mitigate Hallucination
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917731718135808 |
|---|---|
| author | Kollias, Georgios Das, Payel Chaudhury, Subhajit |
| author_facet | Kollias, Georgios Das, Payel Chaudhury, Subhajit |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_16908 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Generation Constraint Scaling Can Mitigate Hallucination Kollias, Georgios Das, Payel Chaudhury, Subhajit Computation and Language Artificial Intelligence Machine Learning 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. |
| title | Generation Constraint Scaling Can Mitigate Hallucination |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2407.16908 |