ArchSpace：记录、评审和引用 LLM 架构研究的公开平台
ArchSpace: Recording, Reviewing, and Referencing LLM Architecture Research
学术圈的架构实验常常算力不够、基线也对不上，这个平台用 scaling laws 让小实验先预测大模型的表现，还直接给了 Qwen3.5 和 OLMo3 基线。做模型架构研究的同学可以上去提交提案或复现别人被验证过的想法，不用每次都从头造轮子。
arXiv 上发布的 ArchSpace 是一个面向 LLM 模型架构研究的公开平台，核心机制是记录（Recording）、评审（Reviewing）和引用（Referencing）：每个架构提案连同实验和推理过程一起被记录，由社区评审，并被后续工作引用。
针对学术研究受限于算力和基线不一致、只能做小规模单点比较的问题，ArchSpace 利用 scaling laws 从小规模实验预测更大规模下的模型 loss，这与 DeepSeek、Qwen、Kimi 等工业界团队的做法一致，让想法在大规模训练前先得到低成本验证。
平台首批提供了 Qwen3.5 和 OLMo3 的基线、拟合好的 scaling laws 以及可复用的训练配方，通过评审验证的想法可以成为新的基线，供下一代模型的设计参考。
ArchSpace: Recording, Reviewing, and Referencing LLM Architecture Research
Model architecture shapes the scaling behavior, training efficiency, and inference cost of LLMs, making it a central focus of both industry and academia. In industry, teams such as those behind DeepSeek, Qwen, and Kimi routinely introduce architectural innovations, yet public reports rarely document which ideas succeeded, which were abandoned, and why. In academia, limited compute and inconsistent baselines often restrict evaluations to small-scale, single-point comparisons, making it difficult to establish whether improvements hold at larger scales or translate into broader impact. ArchSpace addresses both gaps as a public platform for Recording, Reviewing, and Referencing Architecture Researc}: each proposal is recorded with its experiments and reasoning, reviewed by the community, and referenced by later work. To evaluate ideas before costly large-scale training, ArchSpace leverages scaling laws to predict model loss at larger scales from smaller experiments, a practice widely adopted in industry. We provide initial references that make the platform usable: Qwen3.5 and OLMo3 baselines, fitted scaling laws, and a reusable training recipe. Ideas validated through this review can in turn become new baselines, and ArchSpace thereby aims to broaden the range of ideas that shape the next generation of models.
本条情报归属 AITOP「论文」核心栏目，重点聚焦北美前沿实验室与硅谷科技巨头。在跨区域产业观察坐标中，该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线，反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括：ArchSpace、Qwen3.5、OLMo3、scaling laws，读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。
信源核验说明：本文原始线索采编自「arXiv cs.LG」（@Jie Wang, Tong Zhu, JiaHao Kuang, Zhe Xu, Yunkai Chai, Dehong Zheng, Yu Cheng, Tao Ji, ZhouHan Lin, Qi Zhang, Bowen Zhou），经 AITOP 多维评分管线结构化提炼。