@usmananwar.bsky.social
Transformers are REALLY good at in-context learning (ICL); but do they learn ‘adversarially robust’ ICL algorithms? We study this and much more in our new paper! 🧵
November 11, 2024 at 4:20 PM
Transformers are REALLY good at in-context learning (ICL); but do they learn ‘adversarially robust’ ICL algorithms? We study this and much more in our new paper! 🧵