# 评分参考 · 任务目录

HF / OPENENV · REFERENCE ONLY
评分参考 · 任务目录
题面覆盖样本单列，不进入七模型主对照；来源未给出的模型、提供商和运行编号保持为空。
上游输出，非本馆排名
独立的 HuggingFace / OpenEnv 评分参考，不进入主时间线、全部作品和案例总数。
题面覆盖样本单列，不进入七模型主对照；来源未给出的模型、提供商和运行编号保持为空。
pelican_bicycle · 覆盖样本，模型未公开
上游输出，非本馆排名 · reward 1.000
A cartoon pelican is riding a red bicycle with clouds in the background.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "pelican_bicycle",
  "sample": null,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon pelican is riding a red bicycle with clouds in the background.",
  "latency": null,
  "svg": "svg/coverage__pelican_bicycle.svg",
  "png": "png/coverage__pelican_bicycle.png"
}
pelican_unicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.838
A cartoon bird with a yellow beak and legs is riding a unicycle.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "pelican_unicycle",
  "sample": null,
  "reward": 0.8375,
  "structure_score": 1.0,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon bird with a yellow beak and legs is riding a unicycle.",
  "latency": null,
  "svg": "svg/coverage__pelican_unicycle.svg",
  "png": "png/coverage__pelican_unicycle.png"
}
pelican_tandem-bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.656
Two white birds with orange beaks are riding a red tandem bicycle under a sunny blue sky with clouds and a few birds flying.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "pelican_tandem-bicycle",
  "sample": null,
  "reward": 0.65625,
  "structure_score": 0.7142857142857143,
  "semantic_score": 0.625,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "Two white birds with orange beaks are riding a red tandem bicycle under a sunny blue sky with clouds and a few birds flying.",
  "latency": null,
  "svg": "svg/coverage__pelican_tandem-bicycle.svg",
  "png": "png/coverage__pelican_tandem-bicycle.png"
}
pelican_scooter · benchmark-catalogue
上游输出，非本馆排名 · reward 0.716
A cartoon stork is riding a scooter on a road with grass, clouds, and a sun in the background.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "pelican_scooter",
  "sample": null,
  "reward": 0.715625,
  "structure_score": 1.0,
  "semantic_score": 0.5625,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon stork is riding a scooter on a road with grass, clouds, and a sun in the background.",
  "latency": null,
  "svg": "svg/coverage__pelican_scooter.svg",
  "png": "png/coverage__pelican_scooter.png"
}
pelican_skateboard · benchmark-catalogue
上游输出，非本馆排名 · reward 0.625
A cartoon stork is skateboarding on a road under a sunny sky with clouds.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "pelican_skateboard",
  "sample": null,
  "reward": 0.625,
  "structure_score": 0.8571428571428571,
  "semantic_score": 0.5,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon stork is skateboarding on a road under a sunny sky with clouds.",
  "latency": null,
  "svg": "svg/coverage__pelican_skateboard.svg",
  "png": "png/coverage__pelican_skateboard.png"
}
flamingo_bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 1.000
A pink flamingo is riding a red bicycle under a blue sky with clouds and a sun.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "flamingo_bicycle",
  "sample": null,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A pink flamingo is riding a red bicycle under a blue sky with clouds and a sun.",
  "latency": null,
  "svg": "svg/coverage__flamingo_bicycle.svg",
  "png": "png/coverage__flamingo_bicycle.png"
}
flamingo_unicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 1.000
A pink flamingo is riding a unicycle on grass under a blue sky with clouds and a sun.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "flamingo_unicycle",
  "sample": null,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A pink flamingo is riding a unicycle on grass under a blue sky with clouds and a sun.",
  "latency": null,
  "svg": "svg/coverage__flamingo_unicycle.svg",
  "png": "png/coverage__flamingo_unicycle.png"
}
flamingo_tandem-bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.738
Two pink flamingos are riding a tandem bicycle under a sunny sky with clouds and grass below.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "flamingo_tandem-bicycle",
  "sample": null,
  "reward": 0.7375,
  "structure_score": 0.7142857142857143,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "Two pink flamingos are riding a tandem bicycle under a sunny sky with clouds and grass below.",
  "latency": null,
  "svg": "svg/coverage__flamingo_tandem-bicycle.svg",
  "png": "png/coverage__flamingo_tandem-bicycle.png"
}
flamingo_scooter · benchmark-catalogue
上游输出，非本馆排名 · reward 0.919
A pink flamingo is riding a scooter on grass under a blue sky with clouds and a sun.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "flamingo_scooter",
  "sample": null,
  "reward": 0.91875,
  "structure_score": 1.0,
  "semantic_score": 0.875,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A pink flamingo is riding a scooter on grass under a blue sky with clouds and a sun.",
  "latency": null,
  "svg": "svg/coverage__flamingo_scooter.svg",
  "png": "png/coverage__flamingo_scooter.png"
}
flamingo_skateboard · benchmark-catalogue
上游输出，非本馆排名 · reward 0.900
A pink flamingo is riding on a skateboard under a blue sky with clouds and a yellow sun.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "flamingo_skateboard",
  "sample": null,
  "reward": 0.9,
  "structure_score": 0.7142857142857143,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A pink flamingo is riding on a skateboard under a blue sky with clouds and a yellow sun.",
  "latency": null,
  "svg": "svg/coverage__flamingo_skateboard.svg",
  "png": "png/coverage__flamingo_skateboard.png"
}
capybara_bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.838
A cartoon bear is riding a red bicycle on a green ground with a blue background.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "capybara_bicycle",
  "sample": null,
  "reward": 0.8375,
  "structure_score": 1.0,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon bear is riding a red bicycle on a green ground with a blue background.",
  "latency": null,
  "svg": "svg/coverage__capybara_bicycle.svg",
  "png": "png/coverage__capybara_bicycle.png"
}
capybara_unicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.716
A cartoon bear is riding a unicycle on a green line with motion lines indicating movement.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "capybara_unicycle",
  "sample": null,
  "reward": 0.715625,
  "structure_score": 1.0,
  "semantic_score": 0.5625,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon bear is riding a unicycle on a green line with motion lines indicating movement.",
  "latency": null,
  "svg": "svg/coverage__capybara_unicycle.svg",
  "png": "png/coverage__capybara_unicycle.png"
}
capybara_tandem-bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.738
Two cartoon capybaras are riding a tandem bicycle on a sunny day with clouds and bushes visible.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "capybara_tandem-bicycle",
  "sample": null,
  "reward": 0.7375,
  "structure_score": 0.7142857142857143,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "Two cartoon capybaras are riding a tandem bicycle on a sunny day with clouds and bushes visible.",
  "latency": null,
  "svg": "svg/coverage__capybara_tandem-bicycle.svg",
  "png": "png/coverage__capybara_tandem-bicycle.png"
}
capybara_scooter · benchmark-catalogue
上游输出，非本馆排名 · reward 0.553
A cartoon bear is riding a red bicycle on a road under a blue sky with clouds and a yellow sun.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "capybara_scooter",
  "sample": null,
  "reward": 0.553125,
  "structure_score": 1.0,
  "semantic_score": 0.3125,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon bear is riding a red bicycle on a road under a blue sky with clouds and a yellow sun.",
  "latency": null,
  "svg": "svg/coverage__capybara_scooter.svg",
  "png": "png/coverage__capybara_scooter.png"
}
capybara_skateboard · benchmark-catalogue
上游输出，非本馆排名 · reward 0.638
A cartoon bear is standing on a skateboard under a blue sky with clouds and a yellow sun.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "capybara_skateboard",
  "sample": null,
  "reward": 0.6375000000000001,
  "structure_score": 0.42857142857142855,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon bear is standing on a skateboard under a blue sky with clouds and a yellow sun.",
  "latency": null,
  "svg": "svg/coverage__capybara_skateboard.svg",
  "png": "png/coverage__capybara_skateboard.png"
}
axolotl_bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.838
A pink, cartoonish creature with multiple limbs is riding a red bicycle.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "axolotl_bicycle",
  "sample": null,
  "reward": 0.8375,
  "structure_score": 1.0,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A pink, cartoonish creature with multiple limbs is riding a red bicycle.",
  "latency": null,
  "svg": "svg/coverage__axolotl_bicycle.svg",
  "png": "png/coverage__axolotl_bicycle.png"
}
axolotl_unicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.838
A pink cartoon character with red spikes is riding a unicycle on grass under a blue sky with clouds and stars.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "axolotl_unicycle",
  "sample": null,
  "reward": 0.8375,
  "structure_score": 1.0,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A pink cartoon character with red spikes is riding a unicycle on grass under a blue sky with clouds and stars.",
  "latency": null,
  "svg": "svg/coverage__axolotl_unicycle.svg",
  "png": "png/coverage__axolotl_unicycle.png"
}
axolotl_tandem-bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.738
Two cartoonish creatures with pink and purple bodies are riding a tandem bicycle under a sunny blue sky with clouds.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "axolotl_tandem-bicycle",
  "sample": null,
  "reward": 0.7375,
  "structure_score": 0.7142857142857143,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "Two cartoonish creatures with pink and purple bodies are riding a tandem bicycle under a sunny blue sky with clouds.",
  "latency": null,
  "svg": "svg/coverage__axolotl_tandem-bicycle.svg",
  "png": "png/coverage__axolotl_tandem-bicycle.png"
}
axolotl_scooter · benchmark-catalogue
上游输出，非本馆排名 · reward 0.797
A pink cartoon character with antennae is riding a blue scooter under a sunny sky with clouds and grass below.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "axolotl_scooter",
  "sample": null,
  "reward": 0.796875,
  "structure_score": 1.0,
  "semantic_score": 0.6875,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A pink cartoon character with antennae is riding a blue scooter under a sunny sky with clouds and grass below.",
  "latency": null,
  "svg": "svg/coverage__axolotl_scooter.svg",
  "png": "png/coverage__axolotl_scooter.png"
}
axolotl_skateboard · benchmark-catalogue
上游输出，非本馆排名 · reward 0.788
A pink cartoon character with antennae is sitting on a skateboard under a sunny sky with clouds and trees.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "axolotl_skateboard",
  "sample": null,
  "reward": 0.7875000000000001,
  "structure_score": 0.8571428571428571,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A pink cartoon character with antennae is sitting on a skateboard under a sunny sky with clouds and trees.",
  "latency": null,
  "svg": "svg/coverage__axolotl_skateboard.svg",
  "png": "png/coverage__axolotl_skateboard.png"
}
octopus_bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 1.000
A purple octopus is riding a red bicycle with bubbles floating around.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "octopus_bicycle",
  "sample": null,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A purple octopus is riding a red bicycle with bubbles floating around.",
  "latency": null,
  "svg": "svg/coverage__octopus_bicycle.svg",
  "png": "png/coverage__octopus_bicycle.png"
}
octopus_unicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 1.000
A purple octopus is riding a unicycle against a dark background with small yellow dots.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "octopus_unicycle",
  "sample": null,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A purple octopus is riding a unicycle against a dark background with small yellow dots.",
  "latency": null,
  "svg": "svg/coverage__octopus_unicycle.svg",
  "png": "png/coverage__octopus_unicycle.png"
}
octopus_tandem-bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.838
A pink octopus is riding a red bicycle on a road with a blue sky and clouds in the background.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "octopus_tandem-bicycle",
  "sample": null,
  "reward": 0.8375,
  "structure_score": 1.0,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A pink octopus is riding a red bicycle on a road with a blue sky and clouds in the background.",
  "latency": null,
  "svg": "svg/coverage__octopus_tandem-bicycle.svg",
  "png": "png/coverage__octopus_tandem-bicycle.png"
}
octopus_scooter · benchmark-catalogue
上游输出，非本馆排名 · reward 1.000
A pink octopus is riding a blue scooter with two wheels.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "octopus_scooter",
  "sample": null,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A pink octopus is riding a blue scooter with two wheels.",
  "latency": null,
  "svg": "svg/coverage__octopus_scooter.svg",
  "png": "png/coverage__octopus_scooter.png"
}
octopus_skateboard · benchmark-catalogue
上游输出，非本馆排名 · reward 0.788
A purple octopus with large eyes is holding an orange and yellow striped object against a light blue background.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "octopus_skateboard",
  "sample": null,
  "reward": 0.7875000000000001,
  "structure_score": 0.8571428571428571,
  "semantic_score": 0.75,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A purple octopus with large eyes is holding an orange and yellow striped object against a light blue background.",
  "latency": null,
  "svg": "svg/coverage__octopus_skateboard.svg",
  "png": "png/coverage__octopus_skateboard.png"
}
hedgehog_bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 1.000
A cartoon hedgehog is riding a red bicycle on a road under a blue sky with clouds and a sun.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "hedgehog_bicycle",
  "sample": null,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon hedgehog is riding a red bicycle on a road under a blue sky with clouds and a sun.",
  "latency": null,
  "svg": "svg/coverage__hedgehog_bicycle.svg",
  "png": "png/coverage__hedgehog_bicycle.png"
}
hedgehog_unicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 1.000
A cartoon hedgehog is riding a unicycle on a grassy field with colorful objects floating in the background.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "hedgehog_unicycle",
  "sample": null,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon hedgehog is riding a unicycle on a grassy field with colorful objects floating in the background.",
  "latency": null,
  "svg": "svg/coverage__hedgehog_unicycle.svg",
  "png": "png/coverage__hedgehog_unicycle.png"
}
hedgehog_tandem-bicycle · benchmark-catalogue
上游输出，非本馆排名 · reward 0.362
Two hedgehogs are riding a tandem bicycle on a simple road with clouds in the sky.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "hedgehog_tandem-bicycle",
  "sample": null,
  "reward": 0.36171875,
  "structure_score": 0.7142857142857143,
  "semantic_score": 0.171875,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "Two hedgehogs are riding a tandem bicycle on a simple road with clouds in the sky.",
  "latency": null,
  "svg": "svg/coverage__hedgehog_tandem-bicycle.svg",
  "png": "png/coverage__hedgehog_tandem-bicycle.png"
}
hedgehog_scooter · benchmark-catalogue
上游输出，非本馆排名 · reward 1.000
A cartoon hedgehog wearing a yellow helmet and goggles is riding a scooter with a light blue background and clouds.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "hedgehog_scooter",
  "sample": null,
  "reward": 1.0,
  "structure_score": 1.0,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon hedgehog wearing a yellow helmet and goggles is riding a scooter with a light blue background and clouds.",
  "latency": null,
  "svg": "svg/coverage__hedgehog_scooter.svg",
  "png": "png/coverage__hedgehog_scooter.png"
}
hedgehog_skateboard · benchmark-catalogue
上游输出，非本馆排名 · reward 0.950
A cartoon hedgehog is riding a red skateboard under a blue sky with a yellow sun and white clouds.
上游 SVG ↗
·
本地 SVG
原始字段与盲评描述
{
  "model": null,
  "access": null,
  "provider": null,
  "task": "hedgehog_skateboard",
  "sample": null,
  "reward": 0.95,
  "structure_score": 0.8571428571428571,
  "semantic_score": 1.0,
  "gate_passed": true,
  "violations": null,
  "judge_model": "Qwen/Qwen2.5-VL-72B-Instruct",
  "judge_blind_caption": "A cartoon hedgehog is riding a red skateboard under a blue sky with a yellow sun and white clouds.",
  "latency": null,
  "svg": "svg/coverage__hedgehog_skateboard.svg",
  "png": "png/coverage__hedgehog_skateboard.png"
}
← 返回评分对照