ResearchBenchmarksList operations evaluation on ListOps (5, 14) (test)Follow53.1Mean AccuracyBBT-GRC21.69229.8463846.154May 25, 2026Evaluation ResultsMethodMethodLinksMean AccuracyLongest BinBBT-GRC2026.0553.141.4RIR-GRC2026.0551.638.8MLP-LDRU2026.054941.5LSTM2026.0548.138.9TF (ALiBi)Positional Encoding=ALiBiPositional Encoding=ALiBi2026.0541.725.8TF (NoPE)Positional Encoding=NonePositional Encoding=None2026.0532.524.2TF (Sin.)Positional Encoding=Si...Positional Encoding=Sinusoidal2026.0522.98.7