List your company

Video Question Answering on Video-MME no subtitles v2 (test)

14.91Non-Linear AccuracyQwen3-VL-8B
8.12929.889611.6513.4104May 11, 2026

Evaluation results

MethodLinks
Qwen3-VL-8BPipeline=Monolithic Baselines, K (Frame Pool Size)=144, Model Size=8B2026.0514.9130.942,441
Qwen3-VL-4BPipeline=Monolithic Baselines, K (Frame Pool Size)=144, Model Size=4B2026.0514.1130.061,415
GP-4BPipeline=Single-Model GridProbe, Selector Model=Qwen3-VL-4B, QA Model=Qwen3-VL-4B, K (Frame Pool Size)=1442026.0512.8628.28440
GP-8BPipeline=Single-Model GridProbe, Selector Model=Qwen3-VL-8B, QA Model=Qwen3-VL-8B, K (Frame Pool Size)=1442026.0512.628.06842
GP-2B->8BPipeline=Cross-Model GridProbe, Selector Model=Qwen3-VL-2B, QA Model=Qwen3-VL-8B, K (Frame Pool Size)=1442026.0511.726.72677
Uniform-Meff -> 8BPipeline=Cross-Model GridProbe (Control), Selection=Uniform, QA Model=Qwen3-VL-8B, K (Frame Pool Size)=1442026.0510.8326.22677
GP-2B->4BPipeline=Cross-Model GridProbe, Selector Model=Qwen3-VL-2B, QA Model=Qwen3-VL-4B, K (Frame Pool Size)=1442026.0510.7625.22399
Qwen3-VL-2BPipeline=Monolithic Baselines, K (Frame Pool Size)=144, Model Size=2B2026.059.4523.16820
GP-2BPipeline=Single-Model GridProbe, Selector Model=Qwen3-VL-2B, QA Model=Qwen3-VL-2B, K (Frame Pool Size)=1442026.058.3921.53245