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

Meme Generation on MuSE (test)

62.5BERTScore F1

Gemini-3-Flash

Updated 1mo ago

Evaluation Results

Method	Links
Gemini-3-Flash 2026.06		62.5	46.1
Claude-Sonnet-4.6 2026.06		60.7	43.9
Intent Projection 2026.06		58	37.8
Intent Projection 2026.06		57.2	37.1
InternVL3-8B 2026.06		56	35.7
Qwen3-VL-8B 2026.06		55.1	34.8
InternVL3-8B 2026.06		54.3	34.1
GPT-5.4-mini 2026.06		54.2	36.5
Qwen3-VL-8B 2026.06		53.4	33.3
InternVL3-8B 2026.06		53	32.9
Qwen3-VL-8B 2026.06		52.1	32
Intent Projection 2026.06		51.6	32
InternVL3-8B 2026.06		49.9	29.4
LLaVA-OneVision-7B 2026.06		49.5	30
Qwen3-VL-8B 2026.06		48.5	28.3
InternVL3-8B 2026.06		48.2	27.6
LLaVA-OneVision-7B 2026.06		47.9	28.6
Qwen3-VL-8B 2026.06		46.8	26.2
LLaVA-OneVision-7B 2026.06		46.5	27.4
LLaVA-OneVision-7B 2026.06		43.1	24.1
LLaVA-OneVision-7B 2026.06		41.3	22