Question
Will transformer models be the state-of-the-art on most natural language processing benchmarks on January 1, 2027?
Total Forecasters15
Community Prediction95% (78% - 98%)
Make a Prediction
The attention mechanism in transformer models, which calculates correlations between pairs of tokens, takes quadratic time in the input size, making it a time bottleneck for transformer operations.
Attention mechanisms have incurred 7 years' gradual innovations, showing that the transformer architecture is still evolving and improving.
Authors:
Opened:
Closes:
Scheduled resolution:
Comments
? comments
Authors:
Opened:
Closes:
Scheduled resolution: