WildHallucinations: Evaluating Long-form Factuality in LLMs with Real-World Entity Queries
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916335308505088 |
|---|---|
| author | Zhao, Wenting Goyal, Tanya Chiu, Yu Ying Jiang, Liwei Newman, Benjamin Ravichander, Abhilasha Chandu, Khyathi Bras, Ronan Le Cardie, Claire Deng, Yuntian Choi, Yejin |
| author_facet | Zhao, Wenting Goyal, Tanya Chiu, Yu Ying Jiang, Liwei Newman, Benjamin Ravichander, Abhilasha Chandu, Khyathi Bras, Ronan Le Cardie, Claire Deng, Yuntian Choi, Yejin |
| contents | While hallucinations of large language models (LLMs) prevail as a major challenge, existing evaluation benchmarks on factuality do not cover the diverse domains of knowledge that the real-world users of LLMs seek information about. To bridge this gap, we introduce WildHallucinations, a benchmark that evaluates factuality. It does so by prompting LLMs to generate information about entities mined from user-chatbot conversations in the wild. These generations are then automatically fact-checked against a systematically curated knowledge source collected from web search. Notably, half of these real-world entities do not have associated Wikipedia pages. We evaluate 118,785 generations from 15 LLMs on 7,919 entities. We find that LLMs consistently hallucinate more on entities without Wikipedia pages and exhibit varying hallucination rates across different domains. Finally, given the same base models, adding a retrieval component only slightly reduces hallucinations but does not eliminate hallucinations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_17468 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | WildHallucinations: Evaluating Long-form Factuality in LLMs with Real-World Entity Queries Zhao, Wenting Goyal, Tanya Chiu, Yu Ying Jiang, Liwei Newman, Benjamin Ravichander, Abhilasha Chandu, Khyathi Bras, Ronan Le Cardie, Claire Deng, Yuntian Choi, Yejin Computation and Language Artificial Intelligence While hallucinations of large language models (LLMs) prevail as a major challenge, existing evaluation benchmarks on factuality do not cover the diverse domains of knowledge that the real-world users of LLMs seek information about. To bridge this gap, we introduce WildHallucinations, a benchmark that evaluates factuality. It does so by prompting LLMs to generate information about entities mined from user-chatbot conversations in the wild. These generations are then automatically fact-checked against a systematically curated knowledge source collected from web search. Notably, half of these real-world entities do not have associated Wikipedia pages. We evaluate 118,785 generations from 15 LLMs on 7,919 entities. We find that LLMs consistently hallucinate more on entities without Wikipedia pages and exhibit varying hallucination rates across different domains. Finally, given the same base models, adding a retrieval component only slightly reduces hallucinations but does not eliminate hallucinations. |
| title | WildHallucinations: Evaluating Long-form Factuality in LLMs with Real-World Entity Queries |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2407.17468 |