Metraipolis › archive › dc_46bc35b577fd

Understanding LLM Hallucinations Caused by Soft Errors — arXiv:2610.22520

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.22520
License (as archived): CC0-1.0
Archived: 2026-10-19T08:44:23.090Z
SHA-256 of the archived text: 04047c84278db7b2c9c25629e7a7c0f408d9628a6956974d0cfd67c7412246e0
Stable URL of this page: https://metraipolis.com/documento/dc_46bc35b577fd

How to cite

Understanding LLM Hallucinations Caused by Soft Errors — arXiv:2610.22520 — arXiv (en), https://arxiv.org/abs/2610.22520 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_46bc35b577fd (Metraipolis archive dc_46bc35b577fd, sha256 04047c84278db7b2c9c25629e7a7c0f408d9628a6956974d0cfd67c7412246e0)

BibTeX

@misc{dc46bc35b577fd,
  title = {Understanding LLM Hallucinations Caused by Soft Errors — arXiv:2610.22520},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.22520},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_46bc35b577fd; sha256 04047c84278db7b2c9c25629e7a7c0f408d9628a6956974d0cfd67c7412246e0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22520. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

While hallucinations in Large Language Models (LLMs) have been extensively studied from a software perspective, limited attention has been paid to hardware-level factors. In this paper, we performed a case study to understand LLM hallucination caused by soft errors on hardware. We introduce LLMEI to inject soft errors into LLMs during inference, based on what we evaluate the impact of soft error across diverse LLM architectures and tasks, including mathematical reasoning and code generation. We summarize the taxonomy of different failure modes based on LLM hallucination behaviors and analyze their distributions under different error rates. Our find- ings underscore the hallucination-specific diagnostics within the reliability of LLM implementation on hardware.

Tema degli esploratori: calcolo e infrastrutture (elenco arXiv cs.AR, nuovi lavori del giorno).
Categorie arXiv: Software Engineering (cs.SE) ; Hardware Architecture (cs.AR).
Pagina del lavoro: https://arxiv.org/abs/2610.22520 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T08:39:54.825Z dalla pagina https://arxiv.org/list/cs.AR/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).