A learned router sends about 90% of tokens through a linear projection path instead of quadratic attention, keeps the MLP for every token, and matches dense Transformer accuracy at roughly 85% of the FLOPs.
Optimal Design of Climate Disclosure Policies: Transparency versus Externality
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Does a more transparent climate disclosure policy induce lower emissions? This paper examines the welfare implications of transparency in climate disclosure regulation. Increased disclosure transparency could result in a larger equilibrium externality, but never leaves the firm worse off. Consequently, mandating full disclosure is no different from maximizing the firm's private benefit while disregarding the ensuing externality. Transparency beyond binary disclosure is necessary only when the firm holds private information about its incentives for emission reduction. I provide conditions under which focusing on threshold disclosure policies entails no loss of generality.
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DTRNet: Dynamic Token Routing Network to Reduce Quadratic Costs in Transformers
A learned router sends about 90% of tokens through a linear projection path instead of quadratic attention, keeps the MLP for every token, and matches dense Transformer accuracy at roughly 85% of the FLOPs.