Comment by CuriouslyC
1 month ago
LoRAs are better at steering models to produce correct answers from their data set than imparting new knowledge.
1 month ago
LoRAs are better at steering models to produce correct answers from their data set than imparting new knowledge.
https://arxiv.org/abs/2603.01097
>Overall, our findings position LoRA as the complementary axis of memory alongside RAG and ICL, offering distinct advantages.