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

Triplet Recognition on CholecT50

51.69Accuracy (Triplet)

Surg-R1

Updated 1mo ago

Evaluation Results

Method	Links
Surg-R1 2026.03		51.69	86.5	72.15	57.49	87.77	57.16	29.4	16.17
Surg-R1 2026.03		51.69	-	-	-	-	-	-	-
Gemini-3.0-Pro-Thinking 2026.03		8.2	24.32	52.86	12.76	39.29	28.99	16.29	6.86
Gemini-3.0-Pro 2026.03		8.2	-	-	-	-	-	-	-
Qwen2.5-VL-7B-Surg-CholecT50 2026.03		8.01	54.75	43.67	11.08	42.39	28.96	15.15	5.69
GPT-5.1-Thinking 2026.03		6.77	49.31	45.61	12.91	31.73	22.65	12.54	4.59
GPT-5.1 2026.03		6.77	-	-	-	-	-	-	-
InternVL3-8B 2026.03		2.08	51.24	8.94	26.2	22.35	14.01	9.7	2.39
Gemini-2.0-Flash 2026.03		1.85	15.18	7.44	15.87	23.14	14.45	10.91	2.54
GPT-4o-2024-0806 2026.03		1.5	13.33	5.89	5.94	22.8	14.34	10.42	2.6
Qwen2.5-VL-72B-Instruct 2026.03		1.27	32.66	7.91	5.25	23.38	14.45	10.86	2.71
Qwen2.5-VL-32B-Instruct 2026.03		0.98	27.81	7.91	3.12	22.79	14.68	10.09	2.53
Llama-4-Scout-17B-16E-Instruct 2026.03		0.58	9.46	6.17	9.64	22.73	14.22	9.64	2.37
InternVL3-78B 2026.03		0.52	38.14	9.52	3.29	22.74	14.08	9.44	2.36
Mistral-Small-3.1-24B-Instruct-2503 2026.03		0.46	9.81	8.02	5.48	22.43	14.05	9.41	2.35
Qwen2.5-VL-7B-Instruct 2026.03		0.35	10.21	4.9	5.54	22.67	14.14	9.47	2.36
Qwen2.5-Max 2026.03		0.35	7.21	4.85	5.94	22.4	14.12	9.59	2.39
Phi-4-Multimodal 2026.03		0.12	14.83	12.98	3.46	22.32	13.99	9.39	2.35
MiniCPM-O-2_6 2026.03		0.06	24.99	7.79	5.71	22.49	14.02	9.4	2.35
Gemma3-27B-it 2026.03		0.06	8.89	7.39	1.56	22.44	14.01	9.44	2.35
LLaVA-1.5-7B 2026.03		0	3.69	9.41	0.23	22.48	14.12	9.41	2.35
MiniCPM-V-2_6 2026.03		0	18.7	13.04	1.62	22.54	14.2	9.42	2.35
Skywork-R1V-38B 2026.03		0	19.85	9.69	1.67	22.42	14.02	9.41	2.35