arXiv:2606.12342v1 Announce Type: cross Abstract: Domain fine-tuning degrades the safety of large language models: fine-tuned specialists readily comply with harmful prompts framed in domain language. Existing inference-time defenses that mix logits from a safe anchor model require both models to share a vocabulary, which rules them out for the...
Les hele artikkelen hos kilden.
![[Ekstra] Sopra Steria: Hun er ny leder](https://images.gfx.no/1000/2928/2928034/Kristine%20Bolstad%20og%20Kjell%20Rusti.jpg)
Kommentarer (0)
Ingen kommentarer ennå. Bli den første til å kommentere!