KuangshiAi commited on
Commit ·
b7f8c04
1
Parent(s): e7f5b53
adjust task description and rubrics for carp case
Browse files
eval_cases/paraview/paraview_cases.yaml
CHANGED
|
@@ -148,7 +148,7 @@
|
|
| 148 |
|
| 149 |
2. Apply volume rendering to visualize the carp skeleton.
|
| 150 |
|
| 151 |
-
3. Adjust the transfer function to highlight only the bony structures
|
| 152 |
|
| 153 |
4. Optimize the viewpoint to display the full skeleton, ensuring the head, spine, and fins are all clearly visible in a single frame.
|
| 154 |
|
|
@@ -179,13 +179,11 @@
|
|
| 179 |
- type: llm-rubric
|
| 180 |
subtype: vision
|
| 181 |
value: |
|
| 182 |
-
|
| 183 |
|
| 184 |
-
|
| 185 |
|
| 186 |
-
|
| 187 |
-
|
| 188 |
-
4. X-ray Appearance: Does the visualization have a monochrome/grayscale X-ray appearance similar to the ground truth?
|
| 189 |
- type: llm-rubric
|
| 190 |
subtype: text
|
| 191 |
value: |
|
|
|
|
| 148 |
|
| 149 |
2. Apply volume rendering to visualize the carp skeleton.
|
| 150 |
|
| 151 |
+
3. Adjust the transfer function to highlight only the bony structures with the original bone color.
|
| 152 |
|
| 153 |
4. Optimize the viewpoint to display the full skeleton, ensuring the head, spine, and fins are all clearly visible in a single frame.
|
| 154 |
|
|
|
|
| 179 |
- type: llm-rubric
|
| 180 |
subtype: vision
|
| 181 |
value: |
|
| 182 |
+
1. Overall Visualization Goal: Does the result match the ground truth visualization of the carp skeleton?
|
| 183 |
|
| 184 |
+
2. Bone Visibility: Are the bones clearly visible, similar to the ground truth? Are thin fin rays distinguishable?
|
| 185 |
|
| 186 |
+
3. Skeletal Structure: Is the entire carp skeleton (head, spine, ribs, fins, tail) visible and similar in appearance to the ground truth?
|
|
|
|
|
|
|
| 187 |
- type: llm-rubric
|
| 188 |
subtype: text
|
| 189 |
value: |
|
paraview/carp/task_description.txt
CHANGED
|
@@ -13,7 +13,7 @@ Instructions:
|
|
| 13 |
|
| 14 |
2. Apply volume rendering to visualize the carp skeleton.
|
| 15 |
|
| 16 |
-
3. Adjust the transfer function to highlight only the bony structures
|
| 17 |
|
| 18 |
4. Optimize the viewpoint to display the full skeleton, ensuring the head, spine, and fins are all clearly visible in a single frame.
|
| 19 |
|
|
|
|
| 13 |
|
| 14 |
2. Apply volume rendering to visualize the carp skeleton.
|
| 15 |
|
| 16 |
+
3. Adjust the transfer function to highlight only the bony structures with the original bone color.
|
| 17 |
|
| 18 |
4. Optimize the viewpoint to display the full skeleton, ensuring the head, spine, and fins are all clearly visible in a single frame.
|
| 19 |
|
paraview/carp/visualization_goals.txt
CHANGED
|
@@ -1,12 +1,10 @@
|
|
| 1 |
vision:
|
| 2 |
-
1. Overall Visualization Goal: Does the result match the ground truth
|
| 3 |
|
| 4 |
-
2. Bone Visibility: Are the bones clearly visible
|
| 5 |
|
| 6 |
3. Skeletal Structure: Is the entire carp skeleton (head, spine, ribs, fins, tail) visible and similar in appearance to the ground truth?
|
| 7 |
|
| 8 |
-
4. X-ray Appearance: Does the visualization have a monochrome/grayscale X-ray appearance similar to the ground truth?
|
| 9 |
-
|
| 10 |
text:
|
| 11 |
1. Q1 correct answer: C. 7 fins: 1 dorsal, 2 pectoral, 2 pelvic, 1 anal, 1 caudal
|
| 12 |
|
|
|
|
| 1 |
vision:
|
| 2 |
+
1. Overall Visualization Goal: Does the result match the ground truth visualization of the carp skeleton?
|
| 3 |
|
| 4 |
+
2. Bone Visibility: Are the bones clearly visible, similar to the ground truth? Are thin fin rays distinguishable?
|
| 5 |
|
| 6 |
3. Skeletal Structure: Is the entire carp skeleton (head, spine, ribs, fins, tail) visible and similar in appearance to the ground truth?
|
| 7 |
|
|
|
|
|
|
|
| 8 |
text:
|
| 9 |
1. Q1 correct answer: C. 7 fins: 1 dorsal, 2 pectoral, 2 pelvic, 1 anal, 1 caudal
|
| 10 |
|