E-commerce Short Video Understanding on E-VAds 1.0 (test) benchmark leaderboard · SOTA2 Research

E-commerce Short Video Understanding on E-VAds 1.0 (test)

78.6S-BPExpert*

Evaluation Results

Method	Links
Expert*2026.02		78.6	53.6	17.3	25	80.4	53.5	89.3	76.8	58.7	62.5	90.2	76.7	93.2	83.9	79.3	81.3	95.1	87.1
Gemini3-Flash2026.02		76.3	39.1	17.9	19.3	22.2	35	83.6	60.2	58.3	58.8	45.2	61.2	88.8	73	77.6	77.6	63.4	76.1
Qwen3-Omni-Instruct2026.02		69.3	22.7	3.4	3.2	3.7	20.5	80.1	45.4	50.8	50.2	30.6	51.4	86.4	59.6	70.1	69.2	51.2	67.3
GPT 5.22026.02		68.7	31.5	20.5	7.4	10.5	27.8	77.3	52.6	59.6	52.8	26.8	53.9	84.2	67.1	78.6	74.6	47.7	70.4
Qwen3-VL-Instruct2026.02		67.2	23.6	2.3	2.2	2.7	19.6	78.3	46.7	50.1	49.6	25.3	50	85.2	60.9	68.9	68	47.2	66
Qwen3-Omni-Thinking2026.02		63.7	21.4	6.5	5.4	6.5	20.7	75.5	37.9	52.5	51.7	31.5	49.8	83.5	55.2	73.9	72.1	52.8	67.5
E-VAds-R1(Qwen3-VL)2026.02		62.8	25.2	31.3	12.6	27.9	32	74.1	42.7	65.1	55.5	51.5	57.7	82.4	59.2	81.6	76.1	68.9	73.6
Keye1.5-VL2026.02		61.9	12.5	19.3	5.1	3.9	20.6	72.6	55.2	37.7	24.5	50.6	48.2	80.5	75.6	53.2	46.3	69.5	65
Qwen3-VL-Thinking2026.02		60.5	17.2	5.1	5.1	5.1	18.6	72.3	34.9	51.9	51.4	30.7	48.2	81	—	73.8	71.7	51.4	66
Mimo-VL-RL2026.02		60	14.6	0.9	0.7	1	15.5	73.4	34.9	49.1	49	25.8	46.4	81.4	51.4	69.8	64.8	46.3	62.8
Qwen3-VL-Instruct2026.02		60	13.3	0.6	1	1.7	15.3	74.5	36.5	49	49	32.9	48.4	82.8	51.2	62.3	64.4	51.1	62.3
E-VAds-R1(Qwen2.5-VL)2026.02		56.9	12.7	6.9	2.1	17.7	19.3	71.4	31.5	52.3	49.6	45.7	50.1	80.4	49.7	73.9	71.1	65.2	68
Qwen3-VL-Thinking2026.02		56.6	14.4	1.1	0.7	1.2	14.8	71.1	34.2	49.7	49	26.2	46	80.3	51.7	71.7	66.1	47.3	63.4
Qwen2.5-VL-Instruct2026.02		55	15	0.1	0.3	0.3	14.2	70.4	35.4	48.5	49	28.9	46.4	79.9	52.3	69.5	68	47.8	63.5
InternVL32026.02		53.1	9.3	0	0.1	0.2	12.6	67.4	26.6	39.6	41	15.3	38	75.7	42	47.8	47.8	34.3	49.5
InternVL3.52026.02		52.2	8.4	0.4	0.5	0.3	12.4	66.4	26.8	48.3	48	14.6	40.8	75.9	42.3	62.9	62.1	34.7	55.6
Mimo-VL-SFT2026.02		51.9	0.7	12.4	0.8	0.6	13.3	66.4	48.6	30	24.7	47.9	43.5	76.3	67.6	47.1	44.2	61.6	59.4
Qwen2.5-VL-Instruct2026.02		44.6	10.8	0	0	0.1	11.1	60.8	26.8	37.4	39	20.6	36.9	71.4	43.7	45.8	46.8	36.8	48.9
Keye-VL2026.02		30.7	8	6.8	3.8	7.5	11.9	40.8	50.9	18.2	24.3	49.5	37.3	56.8	71.2	38.6	46.2	70.7	57.2
VideoLlama32026.02		30.6	8.2	0	0	0	7.8	44.8	20.8	25.9	28.6	4.9	25	57.8	38.2	39.7	40.4	23.2	39.9