graphslop

An evidence-first failure catalog for GraphRAG and graph-AI claims that sound strong, but collapse under inspection.

taxonomy

Common failure modes in graph extraction and GraphRAG systems. Each category is meant to support evidence-backed critique, not vibes.

Loading patterns...

If this never updates, open the site over HTTP(s) so data can load.

featured entry

One highlighted critique entry, rotated by calendar day (UTC).

failure catalog

Curated entries with source-backed claims, categories, and takeaways.

method

What better graph systems usually do instead, and how this catalog keeps critique disciplined.

Structure first, triples second

Parse documents into logical units (sections, tables, lists) before asking an LLM to assert relations. Text locality matters — fake locality produces fake edges.

Entity resolution is not optional

Canonical IDs, blocking, pairwise scoring, and human review for long tails. Lowercasing strings is not a merge strategy.

Schema as a contract

Version your ontology, constrain predicates, and test extraction against gold sets. If the model drifts types every deploy, you don't have a graph.

Evaluation that matters

Edge-level precision/recall, constraint violations, and retrieval quality beats "the graph looks big." Big graphs full of slop are just high-entropy compost.

Provenance everywhere

Store source spans, document IDs, and extraction version. When someone asks why an edge exists, the answer should not be vibes.

Hybrid pipelines

Rules and dictionaries for stable relations; LLMs for fuzzy bits — with calibration and abstention when confidence is low.