List your company

Question Answering on GPQA-D (∆Acc, CR)

100Coverage RateVanilla
29.2847.646684.36Apr 8, 2026

Evaluation results

MethodLinks
VanillaAccess=—, Base Model=Qwen3-14B (DEER) / Qwen3-8B-Think (ours)2026.04100—
VanillaAccess=—, Base Model=QwQ-32B (DEER) / Qwen3-32B-Think (ours)2026.04100—
DEERAccess=semi-white, Base Model=QwQ-32B (DEER) / Qwen3-32B-Think (ours)2026.04840.7
DEER-PRoAccess=semi-white, Base Model=QwQ-32B (DEER) / Qwen3-32B-Think (ours)2026.04840.9
TCCAccess=black-box, Base Model=Qwen3-14B (DEER) / Qwen3-8B-Think (ours)2026.04610.5
DEER-PRoAccess=semi-white, Base Model=Qwen3-14B (DEER) / Qwen3-8B-Think (ours)2026.04550.9
BAEEAccess=black-box, Base Model=Qwen3-14B (DEER) / Qwen3-8B-Think (ours)2026.04512
Dynasor-CoTAccess=black-box, Base Model=Qwen3-14B (DEER) / Qwen3-8B-Think (ours)2026.04400.4
DEERAccess=semi-white, Base Model=Qwen3-14B (DEER) / Qwen3-8B-Think (ours)2026.04400.8
BAEEAccess=black-box, Base Model=QwQ-32B (DEER) / Qwen3-32B-Think (ours)2026.04331
NoThinkingAccess=black-box, Base Model=Qwen3-14B (DEER) / Qwen3-8B-Think (ours)2026.04329.5