评分 6 · 来源：cs.AI updates on arXiv.org · 发布于 2026-04-29
评分依据：Converts flat SAE features into structured domain-filtered KGs, bridges mechanistic interpretability and structured knowledge
Domain-Filtered Knowledge Graphs from Sparse Autoencoder Features
发布
采集
学术前沿 6.0 分
— Converts flat SAE features into structured domain-filtered KGs, bridges mechanistic interpretability and structured knowledge