Salesforce团队用8B模型生成多样化科研思路超越Claude Opus
Great application of small language models from Salesforce and colleagues.
小模型通过巧妙的奖励机制在创意多样性上击败顶级大模型,展示了高效利用算力的新思路,值得关注。
Great application of small language models from Salesforce and colleagues.
Salesforce及其同事对小语言模型的精彩应用。
AI-scientist pipelines are well known to struggle to propose novel and diverse research ideas.
AI科学家流水线在提出新颖且多样化的研究想法方面一直备受困扰,这一点广为人知。
This team trains an 8B reasoning model to write several research ideas in one generation from a set of source papers.
该团队训练了一个8B推理模型,使其能够从一组源论文中在一次生成中写出多个研究想法。
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.
奖励机制对整个集合进行评分,根据每个想法与之前写出的想法之间的距离来加权其质量。模型学习使每个新想法远离之前的想法。
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.
在其无污染基准测试中,8B模型在创意效用(一种结合质量和多样性的指标)上的得分高于Claude Opus 4.8和GPT-6 Astra。
https://arxiv.org/abs/2610.22324
Chat with Paper: https://academy.dair.ai/papers/training-reasoning-models-to-generate-diverse-scientific-ideas-2610.22324
更进一步:量化金融体系
看懂新闻只是起点——沿量化金融路径,把它变成能交付的工程能力