Salesforce 用 8B 小模型训练 AI 科研创意生成，创意性超 Claude Opus 4.8
Salesforce 只用 8B 小模型就教 AI 一次写出一堆不重复的科研点子，创意性还赢了 Claude Opus 4.8 和 GPT-6 Astra，做法挺巧的。
Salesforce 团队将一个 8B 推理模型训练成一次性生成多个科研创意的 AI-scientist 管线。奖励机制会对整组创意打分，按每个创意与已有创意的距离加权质量，让模型学会拉开新创意与前面创意的差异。在其无数据污染的基准上，这个 8B 模型在衡量质量与多样性的 creative utility 指标上超过 Claude Opus 4.8 和 GPT-6 Astra。论文已发布在 arXiv（编号 2610.22324）。
Great application of small language models from Salesforce and colleagues. AI-scientist pipelines are well known to struggle to propose novel and diverse research ideas. This team trains an 8B reasoning model to write several research ideas in one generation from a set of source papers. The reward scores the whole set, weighting each idea's quality by how far it sits from the ideas written before it. The model learns to move each new idea away from the earlier ones. On their contamination-free benchmark, the 8B model scores higher on creative utility, a measure that combines quality and diversity, than Claude Opus 4.8 and GPT-6 Astra. arxiv.org/abs/2610.22324 Chat with Paper: academy.dair.ai/papers/trainin… 💬 3 🔄 0 ❤️ 5 👀 696 📊 3 ⚡
