List your company

Open-Ended Dialogue (out-of-distribution)

68.3MT-BenchCausalRM
66.11666.68367.2567.817Jan 29, 2026

Evaluation results

MethodLinks
CausalRM2026.0168.360.953.966.262.3
Standard RM2026.0168.257.854.258.559.7
InfoRM2026.0166.760.150.461.959.8
GoalRM2026.0166.259.353.863.660.7