A graph-style RAG helps small models answer cause questions
Small AI models can struggle when they receive several separate text chunks and must work out a long cause-and-effect chain on their own. This approach extracts cause-and-effect links when the data is added, then stores those links as a . When a question is asked, the system returns a connected causal path instead of loose text chunks.
The model then checks and explains the path rather than building the whole chain from scratch. In a 54-question test using for generation, multi-step questions improved from 0.41 with a flat RAG baseline to 0.74 with the causal method. Root-cause questions improved from 0.37 to 0.59, while simple fact lookup stayed about the same.
The query step uses algorithmic retrieval and does not add extra AI calls, which keeps latency low and can reduce token use. The project also includes REST such as /rootcause and /impact.
Key points
- The system extracts cause-and-effect links before questions are asked.
- It stores those links in a and retrieves connected causal paths.
- Multi-step question rose from 0.41 to 0.74 in the reported test.
- Root-cause question rose from 0.37 to 0.59.
- No extra AI calls are used at query time, which can help with latency and token cost.