Share your thoughts, 1 month free Claude Pro on usSee more

Social Reasoning on MMSI

71.2STI

Qwen3.5-9B + SGR

Updated 16d ago

Evaluation Results

Method	Links
Qwen3.5-9B + SGR 2026.05		71.2	66.9	49.2	62.1
Qwen3-VL-8B-Thinking 2026.05		70.4	57	48.8	59
Qwen3.5-9B (Thinking Mode) 2026.05		70.2	61.9	48.2	60
Qwen3.5-9B (Instruct Mode) 2026.05		67.9	68.2	50.1	61.6
Qwen3-VL-8B-Instruct 2026.05		67.5	63.6	48.7	59.7
Qwen3-VL-8B + SGR 2026.05		67.4	63.6	49.8	60.1
LLaVA-OV-1.5-8B-Instruct 2026.05		64.9	65.8	50.9	60.2
LLaVA-OV-1.5-8B-RL 2026.05		63.3	59.8	44.1	55.5
VL-Rethinker-7B 2026.05		62.1	62.7	47.1	56.9
VideoChat-R1.5-7B 2026.05		60.4	65.8	46.9	57.1
Video-R1-7B 2026.05		58.6	60.8	48.8	55.7
Qwen2.5-VL-7B-Instruct 2026.05		57.4	64.1	52.3	57.4
InternVL3.5-8B-Instruct 2026.05		51.5	58.3	50.3	53
LongVILA-R1-7B 2026.05		50.5	44.6	30.5	41.7
InternVL3.5-8B 2026.05		49.2	42.7	37.5	43.2