A transformer layer that iteratively refines its attention matrix to a fixed point is claimed to improve accuracy with no extra parameters, but the benchmark evidence is not reproducible.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Change of Thought: Adaptive Test-Time Computation
A transformer layer that iteratively refines its attention matrix to a fixed point is claimed to improve accuracy with no extra parameters, but the benchmark evidence is not reproducible.