Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Binary Addition on Formal-language benchmark lengths 41-500 (test)
Loading...
100
Accuracy
TRNN
48.312
61.731
75.15
88.569
May 16, 2026
Accuracy
Updated 16d ago
Evaluation Results
Method
Method
Links
Accuracy
TRNN
Architecture=Tape-RNN
2026.05
100
LSTM
2026.05
55.5
Transformer
Architecture=encoder-only
2026.05
54.3
SRNN
Architecture=Stack-RNN
2026.05
52.7
PR-LSTM
2026.05
51.8
RNN
2026.05
50.3
Feedback
Search any
task
Search any
task
