List your company

Independent Scoring on TSQueryBench

0.96Robustness: Linear SpikeQwen 3 8B
0.32560.49030.6550.8197Apr 2, 2026

Evaluation results

MethodLinks
Qwen 3 8BPrompting strategy=rubric-guided, Evaluation protocol=ternary correctness labeling2026.040.96—0.820.750.650.450.910.72
Gemma 2 9BPrompting strategy=rubric-guided, Evaluation protocol=ternary correctness labeling2026.040.58—0.030.680.30.330.330.45
LLaMA 3.1 8BPrompting strategy=rubric-guided, Evaluation protocol=ternary correctness labeling2026.040.35—0.270.220.380.160.270.35