Activation steering with FLORES-derived language vectors produces modest, layer-sensitive and language-dependent gains on cultural awareness tasks, with some settings degrading performance and strong interaction with prompt design.
On the Multilingual Ability of Decoder-based Pre-trained Language Models: Finding and Controlling Language-Specific Neurons , url =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
DFKI-MLT at SemEval-2026 TASK 7: Steering Multilingual Models Towards Cultural Knowledge
Activation steering with FLORES-derived language vectors produces modest, layer-sensitive and language-dependent gains on cultural awareness tasks, with some settings degrading performance and strong interaction with prompt design.