LANCET: Neural Intervention via Structural Entropy for Mitigating Faithfulness Hallucinations in LLMs
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915707449507840 |
|---|---|
| author | Wang, Chenxu Li, Chaozhuo Wang, Pengbo Zhang, Litian Liu, Songyang Qi, Ji Hu, Jiahui Cai, Yushan Zhao, Hao Pu, Rui |
| author_facet | Wang, Chenxu Li, Chaozhuo Wang, Pengbo Zhang, Litian Liu, Songyang Qi, Ji Hu, Jiahui Cai, Yushan Zhao, Hao Pu, Rui |
| contents | Large Language Models have revolutionized information processing, yet their reliability is severely compromised by faithfulness hallucinations. While current approaches attempt to mitigate this issue through node-level adjustments or coarse suppression, they often overlook the distributed nature of neural information, leading to imprecise interventions. Recognizing that hallucinations propagate through specific forward transmission pathways like an infection, we aim to surgically block this flow using precise structural analysis. To leverage this, we propose Lancet, a novel framework that achieves precise neural intervention by leveraging structural entropy and hallucination difference ratios. Lancet first locates hallucination-prone neurons via gradient-driven contrastive analysis, then maps their propagation pathways by minimizing structural entropy, and finally implements a hierarchical intervention strategy that preserves general model capabilities. Comprehensive evaluations across hallucination benchmark datasets demonstrate that Lancet significantly outperforms state-of-the-art methods, validating the effectiveness of our surgical approach to neural intervention. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_01401 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | LANCET: Neural Intervention via Structural Entropy for Mitigating Faithfulness Hallucinations in LLMs Wang, Chenxu Li, Chaozhuo Wang, Pengbo Zhang, Litian Liu, Songyang Qi, Ji Hu, Jiahui Cai, Yushan Zhao, Hao Pu, Rui Computation and Language Machine Learning Large Language Models have revolutionized information processing, yet their reliability is severely compromised by faithfulness hallucinations. While current approaches attempt to mitigate this issue through node-level adjustments or coarse suppression, they often overlook the distributed nature of neural information, leading to imprecise interventions. Recognizing that hallucinations propagate through specific forward transmission pathways like an infection, we aim to surgically block this flow using precise structural analysis. To leverage this, we propose Lancet, a novel framework that achieves precise neural intervention by leveraging structural entropy and hallucination difference ratios. Lancet first locates hallucination-prone neurons via gradient-driven contrastive analysis, then maps their propagation pathways by minimizing structural entropy, and finally implements a hierarchical intervention strategy that preserves general model capabilities. Comprehensive evaluations across hallucination benchmark datasets demonstrate that Lancet significantly outperforms state-of-the-art methods, validating the effectiveness of our surgical approach to neural intervention. |
| title | LANCET: Neural Intervention via Structural Entropy for Mitigating Faithfulness Hallucinations in LLMs |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2601.01401 |