Position paper claiming that AI safety requires explicit runtime controllability and introducing ControlBench to demonstrate gaps in existing alignment methods.
SAE-SSV: Supervised steering in sparse representation spaces for reliable control of language models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Position: AI Safety Requires Effective Controllability
Position paper claiming that AI safety requires explicit runtime controllability and introducing ControlBench to demonstrate gaps in existing alignment methods.