TSA adds end-to-end differentiable per-token halting gates to transformers, enabling learned adaptive depth that saves 14-23% token-layer operations with under 0.5% quality loss on language modeling.
Proceedings of the Annual Meeting of the Association for Computational Linguistics (ACL) , year =
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Adaptive Computation Depth via Learned Token Routing in Transformers
TSA adds end-to-end differentiable per-token halting gates to transformers, enabling learned adaptive depth that saves 14-23% token-layer operations with under 0.5% quality loss on language modeling.