Tiered Language Models use a secret key to induce an alternative computation graph over shared weights, enabling private capabilities in the keyed mode while the public mode shows none.
Findings of the Association for Computational Linguistics: NAACL 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Toward Open Weight Models Without Risks: Separating Public and Private Capabilities in LLMs
Tiered Language Models use a secret key to induce an alternative computation graph over shared weights, enabling private capabilities in the keyed mode while the public mode shows none.