synthetic-disagreement is an open-source Python package for generating deterministic synthetic reviewer disagreement and inter-annotator agreement stress curves. It aids researchers in evaluating annotation quality and simulating disagreement scenarios for robust evaluation.
In the LLM eval & observability space, synthetic-disagreement takes a focused approach. It helps researchers simulate and analyze reviewer disagreement and annotation quality in evaluation workflows. It is built as an open-source project for AI evaluation researchers and data scientists. The project is open source (MIT). synthetic-disagreement is available on the command line, and it can be self-hosted.
It is developed by auraoneai, and it first shipped in 2026. Development happens publicly on GitHub with 12 commits in the last 90 days. Key capabilities include reviewer disagreement simulation, inter-annotator agreement, and stress curve generation.
Summary written by a language model from the project’s public pages.
What PulseGate has recorded for this listing
Same category — not a similarity match