AI 精选动态
智能评分 60
SkyJM-Gen-9B 视觉奖励模型发布
AI 推荐理由
与常见单一分数奖励模型不同，该模型引入结构化rubric评估，值得关注其训练方法对偏好对齐的改进。核心解读
Skywork发布9B参数视觉奖励模型SkyJM-Gen-9B，在MMRB2、GenAI-Bench、GenAI-Bench-Verified上达到72.0/74.1/84.5，采用RubricRM工作流（生成逐提示的评估维度、权重和评分描述）和维度级GRPO训练，Apache 2.0许可。
全文
SkyJM-Gen-9B is now live on ModelScope! A 9B visual reward model for text-to-image generation that ranks candidate images with prompt-conditioned rubrics, not a single flat score.🚀
🤖 https://t.co/HNz4f4U5BH
🏆 Text-to-image judging: tops listed reward models on MMRB2, GenAI-Bench, and GenAI-Bench-Verified, with 72.0 / 74.1 / 84.5
🧩 RubricRM workflow: generates evaluation dimensions, weights, and scoring descriptors for each prompt, then scores both candidate images by dimension
⚙️ Training recipe: rubric-trajectory SFT + dimension-level GRPO for more structured preference judgment
Apache 2.0. vLLM and Transformers ready.
