update readme

README.md

@@ -1,20 +1,33 @@
 ---
-license: cc-by-nc-4.0
+annotations_creators: []
+language_creators:
+- found
+languages:
+- Python, Java
+licenses:
+- cc-by-nc-nd-4.0
+multilinguality:
+- multilingual
+pretty_name: RepoBench-Retrieval
+size_categories:
+- unknown
+source_datasets:
+- original
+task_categories:
+- text-retrieval
+task_ids:
+- document-retrieval
 ---
 
-# Dataset Card for [Dataset Name]
+# Dataset Card for RepoBench-R
 
 ## Table of Contents
-- [Dataset Card for \[Dataset Name\]](#dataset-card-for-dataset-name)
+- [Dataset Card for RepoBench-R](#dataset-card-for-repobench-r)
   - [Table of Contents](#table-of-contents)
   - [Dataset Description](#dataset-description)
     - [Dataset Summary](#dataset-summary)
-    - [Supported Tasks and Leaderboards](#supported-tasks-and-leaderboards)
-    - [Languages](#languages)
+    - [Supported Tasks](#supported-tasks)
   - [Dataset Structure](#dataset-structure)
-    - [Data Instances](#data-instances)
-    - [Data Fields](#data-fields)
-    - [Data Splits](#data-splits)
   - [Dataset Creation](#dataset-creation)
     - [Curation Rationale](#curation-rationale)
     - [Source Data](#source-data)
@@ -36,36 +49,44 @@ license: cc-by-nc-4.0
 
 ## Dataset Description
 
-- **Homepage:**
-- **Paper:**
-- **Leaderboard:**
-- **Point of Contact:**
+- **Homepage:** https://github.com/Leolty/repobench
+- **Paper:** https://arxiv.org/abs/2306.03091
+- **Point of Contact:** til040@ucsd.edu
 
 ### Dataset Summary
 
-[More Information Needed]
+RepoBench-R is a subtask of [RepoBench](https://github.com/Leolty/repobench), targeting the retrieval component of a repository-level auto-completion
+system, focusing on extracting the most relevant code snippet from a project repository for next-line
+code prediction. 
 
-### Supported Tasks and Leaderboards
+### Supported Tasks 
 
-[More Information Needed]
+The dataset supports two programming languages, Python and Java, and contains two settings:
 
-### Languages
+- `cff`: short for `cross_file_first`, where the cross-file module in next line is first used in the current file.
+- `cfr`: short for `cross_file_random`, where the cross-file module in next line is not first used in the current file.
 
-[More Information Needed]
+For each setting, we provide two levels of difficulty: `easy` and `hard`. Suppose the number of code snippets in the context is $k$, then for the *easy* subset, we have $5 \leq k < 10$, while for the *hard* subset, we have $k \geq 10$.
 
-## Dataset Structure
+So, the specific information of each split is as follows:
 
-### Data Instances
+- `python_cff`:
+  - `easy`: 8,000 samples
+  - `hard`: 4,000 samples
+- `python_cfr`:
+  - `easy`: 8,000 samples
+  - `hard`: 4,000 samples
+- `java_cff`:
+  - `easy`: 8,000 samples
+  - `hard`: 4,000 samples
+- `java_cfr`:
+  - `easy`: 8,000 samples
+  - `hard`: 4,000 samples
 
-[More Information Needed]
 
-### Data Fields
-
-[More Information Needed]
+## Dataset Structure
 
-### Data Splits
 
-[More Information Needed]
 
 ## Dataset Creation