Update README.md

README.md

@@ -2,46 +2,47 @@
 pretty_name: CARE-Bench
 
 configs:
-  - config_name: model_inputs
-    default: true
-    data_files:
-      - split: development
-        path: model_inputs_development.jsonl
-      - split: public_test_1
-        path: model_inputs_public_test_1.jsonl
-      - split: validation
-        path: model_inputs_validation.jsonl
+- config_name: model_inputs
+  default: true
+  data_files:
+  - split: development
+    path: "model_inputs_development.jsonl"
+  - split: validation
+    path: "model_inputs_validation.jsonl"
+  - split: public_test_1
+    path: "model_inputs_public_test_1.jsonl"
 
-  - config_name: cases
-    data_files:
-      - split: development
-        path: cases_development.jsonl
-      - split: public_test_1
-        path: cases_public_test_1.jsonl
-      - split: validation
-        path: cases_validation.jsonl
+- config_name: cases
+  data_files:
+  - split: development
+    path: "cases_development.jsonl"
+  - split: validation
+    path: "cases_validation.jsonl"
+  - split: public_test_1
+    path: "cases_public_test_1.jsonl"
 
-  - config_name: evaluation_prefixes
-    data_files:
-      - split: development
-        path: evaluation_prefixes_development.jsonl
-      - split: public_test_1
-        path: evaluation_prefixes_public_test_1.jsonl
-      - split: validation
-        path: evaluation_prefixes_validation.jsonl
+- config_name: evaluation_prefixes
+  data_files:
+  - split: development
+    path: "evaluation_prefixes_development.jsonl"
+  - split: validation
+    path: "evaluation_prefixes_validation.jsonl"
+  - split: public_test_1
+    path: "evaluation_prefixes_public_test_1.jsonl"
 
-  - config_name: submission_template
-    data_files:
-      - split: development
-        path: submission_template_development.csv
-      - split: public_test_1
-        path: submission_template_public_test_1.csv
-      - split: validation
-        path: submission_template_validation.csv
+- config_name: submission_template
+  data_files:
+  - split: development
+    path: "submission_template_development.csv"
+  - split: validation
+    path: "submission_template_validation.csv"
+  - split: public_test_1
+    path: "submission_template_public_test_1.csv"
 ---
+
 # CARE-Bench v1.1 public data
 
-The public release contains 439 cases and 925 labeled prefixes:
+The public release contains 439 cases and 925 labeled prefixes.
 
 | Split | Cases | Prefixes |
 |---|---:|---:|
@@ -58,10 +59,15 @@ The prefix labels are distributed as follows:
 | Nonurgent care | 342 |
 | Urgent care | 172 |
 
-`cases_public.jsonl` contains the released source-grounded cases. It omits internal review annotations, construction-plan fields, and target-round planning notes.
+The Dataset Viewer includes four configurations:
+
+- `model_inputs`: identifiers, split information, and model-facing messages
+- `cases`: released source-grounded cases
+- `evaluation_prefixes`: reference labels and evaluation metadata
+- `submission_template`: submission rows for every released prefix
 
-`evaluation_prefixes_public.jsonl` contains labels and evaluation metadata. `model_inputs_public.jsonl` contains only the identifiers, split, and model-facing messages needed to run a system.
+The `cases` configuration excludes internal review annotations, construction-plan fields, and target-round planning notes.
 
-`submission_template_public.csv` contains every public prefix. The `public_test_1` subset contains 135 scored rows.
+The `public_test_1` split contains 135 scored prefixes.
 
-Controlled-access Test 2 rows and labels are excluded from this directory. Request its input-only data through the [CARE-Bench Test 2 access request form](https://forms.gle/hrmC3ciGcj46f4gW7).
+Controlled-access Test 2 inputs and labels are excluded from the public repository. Input-only data can be requested through the [CARE-Bench Test 2 access request form](https://forms.gle/hrmC3ciGcj46f4gW7).