ArchSpace: Recording, Reviewing, and Referencing LLM Architecture Research
arXiv cs.CL · 13 Nov 2026, 04:59 AEST
ArchSpace is a public platform designed for recording, reviewing, and referencing research on LLM architecture. It aims to bridge gaps in both industry and academia regarding the documentation and evaluation of architectural innovations.
Key facts
- 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.
- ArchSpace addresses both gaps as a public platform for Recording, Reviewing, and Referencing Architecture Research: 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.
Read aloud
ArchSpace is a public platform designed for recording, reviewing, and referencing research on L L M architecture. It aims to bridge gaps in both industry and academia regarding the documentation and evaluation of architectural innovations. Model architecture shapes the scaling behavior, training efficiency, and inference cost of L L M s, 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. ArchSpace addresses both gaps as a public platform for Recording, Reviewing, and Referencing Architecture Research: 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.
Sources in this cluster
- ArchSpace: Recording, Reviewing, and Referencing LLM Architecture Research
arXiv cs.CL · 13 Nov 2026, 04:59 AEST