ResearchBenchmarksList operations evaluation on ListOps (5, 9) (test)Follow49.6Mean AccuracyBBT-GRC20.37627.96335.5543.137May 25, 2026Evaluation ResultsMethodMethodLinksMean AccuracyLongest Bin AccuracyBBT-GRC2026.0549.639.2RIR-GRC2026.0548.136.4MLP-LDRU2026.0545.938.8LSTM2026.0544.936TF (ALiBi)Positional Encoding=ALiBiPositional Encoding=ALiBi2026.053922.8TF (NoPE)Positional Encoding=NonePositional Encoding=None2026.0529.922.7TF (Sin.)Positional Encoding=Si...Positional Encoding=Sinusoidal2026.0521.58.7