Originalarbeiterreichbar
Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?
Gekhman, Yona, Aharoni, Eyal, Feder, Reichart und Herzig (Google, Technion), ar
geprüft 10.11.2026
Worauf sich diese Seite beruft, wörtlich, abgerufen am 21.10.2026:
We demonstrate that large language models struggle to acquire new factual knowledge through fine-tuning, as fine-tuning examples that introduce new knowledge are learned significantly slower than those consistent with the model’s knowledge.
bestätigt 10.11.2026However, we also find that as the examples with new knowledge are eventually learned, they linearly increase the model’s tendency to hallucinate.
bestätigt 10.11.2026Taken together, our results highlight the risk in introducing new factual knowledge through fine-tuning, and support the view that large language models mostly acquire factual knowledge through pre-training, whereas fine-tuning teaches them to use it more efficiently.
bestätigt 10.11.2026
02 Ein Modell auf eigene Daten abstimmen08 Woran es fehlt, entscheidet den Weg