Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2405.06545 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866914790966820864 |
|---|---|
| author | Niu, Mengjia Li, Hao Shi, Jie Haddadi, Hamed Mo, Fan |
| author_facet | Niu, Mengjia Li, Hao Shi, Jie Haddadi, Hamed Mo, Fan |
| contents | Large language models (LLMs) have demonstrated remarkable capabilities across various domains, although their susceptibility to hallucination poses significant challenges for their deployment in critical areas such as healthcare. To address this issue, retrieving relevant facts from knowledge graphs (KGs) is considered a promising method. Existing KG-augmented approaches tend to be resource-intensive, requiring multiple rounds of retrieval and verification for each factoid, which impedes their application in real-world scenarios.
In this study, we propose Self-Refinement-Enhanced Knowledge Graph Retrieval (Re-KGR) to augment the factuality of LLMs' responses with less retrieval efforts in the medical field. Our approach leverages the attribution of next-token predictive probability distributions across different tokens, and various model layers to primarily identify tokens with a high potential for hallucination, reducing verification rounds by refining knowledge triples associated with these tokens. Moreover, we rectify inaccurate content using retrieved knowledge in the post-processing stage, which improves the truthfulness of generated responses. Experimental results on a medical dataset demonstrate that our approach can enhance the factual capability of LLMs across various foundational models as evidenced by the highest scores on truthfulness. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_06545 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Mitigating Hallucinations in Large Language Models via Self-Refinement-Enhanced Knowledge Retrieval Niu, Mengjia Li, Hao Shi, Jie Haddadi, Hamed Mo, Fan Computation and Language Machine Learning I.2.7; H.3.3 Large language models (LLMs) have demonstrated remarkable capabilities across various domains, although their susceptibility to hallucination poses significant challenges for their deployment in critical areas such as healthcare. To address this issue, retrieving relevant facts from knowledge graphs (KGs) is considered a promising method. Existing KG-augmented approaches tend to be resource-intensive, requiring multiple rounds of retrieval and verification for each factoid, which impedes their application in real-world scenarios. In this study, we propose Self-Refinement-Enhanced Knowledge Graph Retrieval (Re-KGR) to augment the factuality of LLMs' responses with less retrieval efforts in the medical field. Our approach leverages the attribution of next-token predictive probability distributions across different tokens, and various model layers to primarily identify tokens with a high potential for hallucination, reducing verification rounds by refining knowledge triples associated with these tokens. Moreover, we rectify inaccurate content using retrieved knowledge in the post-processing stage, which improves the truthfulness of generated responses. Experimental results on a medical dataset demonstrate that our approach can enhance the factual capability of LLMs across various foundational models as evidenced by the highest scores on truthfulness. |
| title | Mitigating Hallucinations in Large Language Models via Self-Refinement-Enhanced Knowledge Retrieval |
| topic | Computation and Language Machine Learning I.2.7; H.3.3 |
| url | https://arxiv.org/abs/2405.06545 |