Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,16 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
---
|
| 4 |
+
|
| 5 |
+
The fine-tuned VLMs and CLIP model used in this work are available at:
|
| 6 |
+
https://huggingface.co/ys-qu/found-rl_vlms
|
| 7 |
+
|
| 8 |
+
The implementation code is available at:
|
| 9 |
+
https://github.com/ys-qu/found-rl
|
| 10 |
+
|
| 11 |
+
We also provide trained RL policy checkpoints for DrQv2-CLIP to facilitate direct evaluation:
|
| 12 |
+
|
| 13 |
+
- `drqv2-clip-lb.zip`: a checkpoint from one of the three random-seed training runs on the Leaderboard benchmark, evaluated on the Leaderboard benchmark.
|
| 14 |
+
- `drqv2-clip-eu.zip`: a checkpoint from one of the three random-seed training runs on the NoCrash benchmark, evaluated under the NoCrash setting.
|
| 15 |
+
|
| 16 |
+
These checkpoints can be used to run evaluation without retraining the RL agents from scratch.
|