List your company

Multiple-choice Question Answering on CSQA

79.85AccuracyK-stage RK
-2.9090418.5764840.06261.54752Feb 23, 2026Mar 9, 2026Mar 24, 2026Apr 8, 2026Apr 22, 2026May 7, 2026May 22, 2026

Evaluation results

MethodLinks
K-stage RKSolver choice=Runge–Kutta, K=22026.0579.85—
Heun K=1Solver choice=Heun, K=12026.0579.61—
Baseline2026.0578.71—
GPT-4o2026.020.7028—
Claude 3 Haiku2026.020.5587—
Llama3-70B-InstructSize=70B, Type=Instruct2026.020.5549—
Mistral Large2026.020.5535—
Qwen1.5 (1.8B)Size=1.8B2026.020.3113—
Gemma (2B)Size=2B2026.020.274—