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

Response Generation on SCREEN (test)

49.5BLEU-1

SiPeR

Updated 1mo ago

Evaluation Results

Method	Links
SiPeR 2026.04		49.5	36.44	45.48	38.5	9.35
w/o BI-INF 2026.04		47.42	35.21	41.88	37.29	8.95
w/o STE 2026.04		46.22	34.45	41.08	36.21	8.78
Qwen2.5-VL 2026.04		45.34	33.9	40.77	35.91	8.52
LLaVA-NeXT 2026.04		43.67	29.92	38.29	33.29	8.24
GPT-4o 2026.04		42.39	31.56	42.04	36.11	8.12
GPT-4o 2026.04		40.52	28.34	40.72	36.21	7.85
ReGeS 2026.04		39.18	31.52	37.45	33.12	8.08
Qwen2.5-VL 2026.04		37.75	26.34	31.58	24.92	6.58
ALBEF 2026.04		34.29	24.12	26.21	20.23	6.92
Qwen2.5-VL 2026.04		34.22	22.85	25.56	19.12	6.25
LLaVA-NeXT 2026.04		33.26	22.9	30.59	23.21	6.42
LLaVA-NeXT 2026.04		28.75	19.68	23.85	18.24	6.12