Commit
•
c9c6b9a
1
Parent(s):
1a5bdd9
link
Browse files- .gitignore +5 -0
- src/assets/text_content.py +3 -1
.gitignore
CHANGED
|
@@ -1,2 +1,7 @@
|
|
| 1 |
*/.DS_Store
|
| 2 |
.DS_Store
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
*/.DS_Store
|
| 2 |
.DS_Store
|
| 3 |
+
|
| 4 |
+
.idea/
|
| 5 |
+
src/__pycache__/
|
| 6 |
+
src/assets/__pycache__/
|
| 7 |
+
venv/
|
src/assets/text_content.py
CHANGED
|
@@ -5,7 +5,9 @@ The CLEM Leaderboard aims to track, rank and evaluate current cLLMs (chat-optimi
|
|
| 5 |
|
| 6 |
The benchmarking approach is described in [Clembench: Using Game Play to Evaluate Chat-Optimized Language Models as Conversational Agents](https://arxiv.org/abs/2305.13455).
|
| 7 |
|
| 8 |
-
[
|
|
|
|
|
|
|
| 9 |
"""
|
| 10 |
|
| 11 |
SHORT_NAMES = {
|
|
|
|
| 5 |
|
| 6 |
The benchmarking approach is described in [Clembench: Using Game Play to Evaluate Chat-Optimized Language Models as Conversational Agents](https://arxiv.org/abs/2305.13455).
|
| 7 |
|
| 8 |
+
Source code benchmarking "clems" is available here: [Clembench](https://github.com/clembench/clembench)
|
| 9 |
+
|
| 10 |
+
All generated files and results from the benchmark runs are available here: [clembench-runs](https://github.com/clembench/clembench-runs)
|
| 11 |
"""
|
| 12 |
|
| 13 |
SHORT_NAMES = {
|