Fine-Grained Self-Endorsement Improves Factuality and Reasoning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929256174452736 |
|---|---|
| author | Wang, Ante Song, Linfeng Peng, Baolin Tian, Ye Jin, Lifeng Mi, Haitao Su, Jinsong Yu, Dong |
| author_facet | Wang, Ante Song, Linfeng Peng, Baolin Tian, Ye Jin, Lifeng Mi, Haitao Su, Jinsong Yu, Dong |
| contents | This work studies improving large language model (LLM) generations at inference time by mitigating fact-conflicting hallucinations. Particularly, we propose a self-endorsement framework that leverages the fine-grained fact-level comparisons across multiple sampled responses. Compared with prior ensemble methods (Wang et al., 2022;Chen et al., 2023)) that perform response-level selection, our approach can better alleviate hallucinations, especially for longform generation tasks. Our approach can broadly benefit smaller and open-source LLMs as it mainly conducts simple content-based comparisons. Experiments on Biographies show that our method can effectively improve the factuality of generations with simple and intuitive prompts across different scales of LLMs. Besides, comprehensive analyses on TriviaQA and GSM8K demonstrate the potential of self-endorsement for broader application. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_15631 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Fine-Grained Self-Endorsement Improves Factuality and Reasoning Wang, Ante Song, Linfeng Peng, Baolin Tian, Ye Jin, Lifeng Mi, Haitao Su, Jinsong Yu, Dong Computation and Language Artificial Intelligence This work studies improving large language model (LLM) generations at inference time by mitigating fact-conflicting hallucinations. Particularly, we propose a self-endorsement framework that leverages the fine-grained fact-level comparisons across multiple sampled responses. Compared with prior ensemble methods (Wang et al., 2022;Chen et al., 2023)) that perform response-level selection, our approach can better alleviate hallucinations, especially for longform generation tasks. Our approach can broadly benefit smaller and open-source LLMs as it mainly conducts simple content-based comparisons. Experiments on Biographies show that our method can effectively improve the factuality of generations with simple and intuitive prompts across different scales of LLMs. Besides, comprehensive analyses on TriviaQA and GSM8K demonstrate the potential of self-endorsement for broader application. |
| title | Fine-Grained Self-Endorsement Improves Factuality and Reasoning |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2402.15631 |