Blog

Notes from the forest

Engineering notes on navigation, ingestion, agent memory and honest benchmarking, written by the people who ran the experiments.

benchmark transparency rag
Inside the 11-Question Benchmark
Eleven questions, each needing at least three chained hops. The same 12B model scores 0/11 as a top-k RAG reader and 11/11 as a forest navigator. Here is how the benchmark was built and how to rerun it.
August 26, 20264 min read
benchmark transparency open-source
Our benchmark has a failure section. On purpose.
The trail-learning convergence criterion was not met, and it is in the report anyway. A benchmark you can only pass is not a benchmark.
August 21, 20263 min read

Want the long version?

The paper carries the full architecture, the benchmark tables and the findings that failed their criteria.