List your company

Mathematical Reasoning on GSM8K (accuracy, speedup)

89.6Accuracy (GSM8K)Diffusion (static)
73.16877.43481.785.966Mar 26, 2026

Evaluation results

MethodLinks
Diffusion (static)Model Scale=SDAR-8B-Chat, Decoding Strategy=Diffusion (static)2026.0389.61.4
S2D2 (config-A)Model Scale=SDAR-8B-Chat, Decoding Strategy=S2D2 (config-A)2026.0389.62
ARModel Scale=SDAR-8B-Chat, Decoding Strategy=AR2026.0389.31
Diffusion (dynamic)Model Scale=SDAR-8B-Chat, Decoding Strategy=Diffusion (dynamic)2026.0389.32.6
Diffusion (static)Model Scale=SDAR-4B-Chat, Decoding Strategy=Diffusion (static)2026.03891.4
Diffusion (dynamic)Model Scale=SDAR-4B-Chat, Decoding Strategy=Diffusion (dynamic)2026.0388.92.7
ARModel Scale=SDAR-4B-Chat, Decoding Strategy=AR2026.0388.51
S2D2 (config-B)Model Scale=SDAR-8B-Chat, Decoding Strategy=S2D2 (config-B)2026.0388.33.8
S2D2 (config-A)Model Scale=SDAR-4B-Chat, Decoding Strategy=S2D2 (config-A)2026.0387.53.7
S2D2 (config-B)Model Scale=SDAR-4B-Chat, Decoding Strategy=S2D2 (config-B)2026.0387.44.3
Diffusion (static)Model Scale=SDAR-1.7B-Chat, Decoding Strategy=Diffusion (static)2026.0377.91.7
S2D2 (config-A)Model Scale=SDAR-1.7B-Chat, Decoding Strategy=S2D2 (config-A)2026.0377.42.3
Diffusion (dynamic)Model Scale=SDAR-1.7B-Chat, Decoding Strategy=Diffusion (dynamic)2026.0377.22.7
ARModel Scale=SDAR-1.7B-Chat, Decoding Strategy=AR2026.0376.31
S2D2 (config-B)Model Scale=SDAR-1.7B-Chat, Decoding Strategy=S2D2 (config-B)2026.0373.84.3