Enhancing Factual Accuracy and Citation Generation in LLMs via Multi-Stage Self-Verification
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908522483023872 |
|---|---|
| author | García, Fernando Gabriela Shi, Qiyang Feng, Zilin |
| author_facet | García, Fernando Gabriela Shi, Qiyang Feng, Zilin |
| contents | This research introduces VeriFact-CoT (Verified Factual Chain-of-Thought), a novel method designed to address the pervasive issues of hallucination and the absence of credible citation sources in Large Language Models (LLMs) when generating complex, fact-sensitive content. By incorporating a multi-stage mechanism of 'fact verification-reflection-citation integration,' VeriFact-CoT empowers LLMs to critically self-examine and revise their intermediate reasoning steps and final answers. This process significantly enhances the objective accuracy, trustworthiness, and traceability of the generated outputs, making LLMs more reliable for applications demanding high fidelity such as scientific research, news reporting, and legal consultation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_05741 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Enhancing Factual Accuracy and Citation Generation in LLMs via Multi-Stage Self-Verification García, Fernando Gabriela Shi, Qiyang Feng, Zilin Computation and Language This research introduces VeriFact-CoT (Verified Factual Chain-of-Thought), a novel method designed to address the pervasive issues of hallucination and the absence of credible citation sources in Large Language Models (LLMs) when generating complex, fact-sensitive content. By incorporating a multi-stage mechanism of 'fact verification-reflection-citation integration,' VeriFact-CoT empowers LLMs to critically self-examine and revise their intermediate reasoning steps and final answers. This process significantly enhances the objective accuracy, trustworthiness, and traceability of the generated outputs, making LLMs more reliable for applications demanding high fidelity such as scientific research, news reporting, and legal consultation. |
| title | Enhancing Factual Accuracy and Citation Generation in LLMs via Multi-Stage Self-Verification |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.05741 |