Describe the benchmark as a test of SystemOne decision models

README.md

@@ -7,9 +7,9 @@ viewer: false
 
 # Perch benchmark results
 
-This benchmark measures how well a model finds bugs and security issues in a single code method, and how often it flags clean code. Each record contains a method before and after a code change, plus input adapted from a [Perch scan](https://github.com/lakeday-org/perch): method source, imports, callers, callees, and static call-graph edges where available. Models score each version independently, without seeing the repair. Benign reference methods are included, so a `before` method is not automatically positive.
+This dataset is for evaluating SystemOne decision models on finding bugs and security issues in a single method. It also measures how often they flag clean code. Decision models differ from generative LLMs: they answer a specific question about the input and output only a probability. Examples are [Jev](https://typesafe.ai/blog/introducing-system-one-models-and-jev), [Liquid d1](https://docs.liquid.ai/lfm/models/decision-models), [OpenAI's Decisions API](https://developers.openai.com/api/docs/guides/decisions) and [GLiNER-2.5-Decide](https://huggingface.co/fastino/GLiNER2.5-Decide). We also run generative LLMs, other classifiers and CodeQL on the same pairs for comparison. Each record contains a method before and after a code change, plus input adapted from a [Perch scan](https://github.com/lakeday-org/perch): method source, imports, callers, callees, and static call-graph edges where available. Models score each version independently, without seeing the repair. Benign reference methods are included, so a `before` method is not automatically positive.
 
-Pairs come from open-source projects in Rust, JavaScript, TypeScript, Python, Go, Java, C, C++, C#, PHP, and other languages. Bug, security, and user-written semantic lint rules are separate tracks. Public and private splits are disjoint by source repository. Any classifier or LLM can be scored with the command adapter under [Files and reproduction](#files-and-reproduction).
+Pairs come from open-source projects in Rust, JavaScript, TypeScript, Python, Go, Java, C, C++, C#, PHP, and other languages. Bug, security, and user-written semantic lint rules are separate tracks. Public and private splits are disjoint by source repository. See [Files and reproduction](#files-and-reproduction) for how to score any model with the command adapter.
 
 ## Inference source
 
@@ -215,4 +215,4 @@ per minute. Recorded method elapsed times include client throttling and rate-lim
 
 ## License
 
-Results, predictions, the question pack and this card: [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Scripts (`*.py`, `*.mjs` and `adapters/`): MIT. The pair data lives in the [bug](https://huggingface.co/datasets/perchscan/bug-pairs-public), [security](https://huggingface.co/datasets/perchscan/security-pairs-public) and [lint](https://huggingface.co/datasets/perchscan/perch-lint-bench-public) datasets, each with its own license and source credits.
+Everything here except the scripts is licensed under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/): the results, predictions, question pack and this card. The scripts (`*.py`, `*.mjs` and `adapters/`) are licensed under MIT. The pair data is in the [bug](https://huggingface.co/datasets/perchscan/bug-pairs-public), [security](https://huggingface.co/datasets/perchscan/security-pairs-public) and [lint](https://huggingface.co/datasets/perchscan/perch-lint-bench-public) datasets, each with its own license. See their cards for the data sources.