docs: document LB-SQAV citation shorthand
Browse files
README.md
CHANGED
|
@@ -28,6 +28,8 @@ configs:
|
|
| 28 |
|
| 29 |
# LexBench-SimpleQA-Verified
|
| 30 |
|
|
|
|
|
|
|
| 31 |
A 831-question retrieval edition of [Google SimpleQA Verified](https://huggingface.co/datasets/google/simpleqa-verified), used by [Browser Eval](https://github.com/lexmount/browser-eval) to compare search tools. It retains publisher questions, reference answers and supporting URLs so a search result list can be checked for reference-page retrieval and answer-bearing text.
|
| 32 |
|
| 33 |
Google DeepMind and Google Research designed the original benchmark to measure a model's factual knowledge without search tools. This derivative changes the task to retrieval evaluation. Its retrieval metrics are not official SimpleQA Verified accuracy or evidence of a model's closed-book factual knowledge.
|
|
|
|
| 28 |
|
| 29 |
# LexBench-SimpleQA-Verified
|
| 30 |
|
| 31 |
+
Short name: **LB-SQAV**.
|
| 32 |
+
|
| 33 |
A 831-question retrieval edition of [Google SimpleQA Verified](https://huggingface.co/datasets/google/simpleqa-verified), used by [Browser Eval](https://github.com/lexmount/browser-eval) to compare search tools. It retains publisher questions, reference answers and supporting URLs so a search result list can be checked for reference-page retrieval and answer-bearing text.
|
| 34 |
|
| 35 |
Google DeepMind and Google Research designed the original benchmark to measure a model's factual knowledge without search tools. This derivative changes the task to retrieval evaluation. Its retrieval metrics are not official SimpleQA Verified accuracy or evidence of a model's closed-book factual knowledge.
|