Spaces:
Running
Running
Commit
·
d0e494a
1
Parent(s):
17ad9a6
updating metrics
Browse files
src/data/metrics/absolute_improvement_to_baseline.json
CHANGED
@@ -54,13 +54,13 @@
|
|
54 |
"MLAB (llama3-1-405b-instruct)": 3.8
|
55 |
},
|
56 |
"erasing_invisible_watermarks": {
|
57 |
-
"CoI-Agent (o1) + MLAB (gpt-4o)":
|
58 |
-
"Top Human in Competition":
|
59 |
-
"MLAB (claude-3-5-sonnet-v2)":
|
60 |
-
"MLAB (gemini-exp-1206)":
|
61 |
-
"MLAB (o3-mini)":
|
62 |
-
"MLAB (gpt-4o)":
|
63 |
-
"MLAB (llama3-1-405b-instruct)":
|
64 |
},
|
65 |
"backdoor-trigger-recovery": {
|
66 |
"CoI-Agent (o1) + MLAB (gpt-4o)": 85.0,
|
|
|
54 |
"MLAB (llama3-1-405b-instruct)": 3.8
|
55 |
},
|
56 |
"erasing_invisible_watermarks": {
|
57 |
+
"CoI-Agent (o1) + MLAB (gpt-4o)": 80.3,
|
58 |
+
"Top Human in Competition": 95.6,
|
59 |
+
"MLAB (claude-3-5-sonnet-v2)": 83.7,
|
60 |
+
"MLAB (gemini-exp-1206)": 93.3,
|
61 |
+
"MLAB (o3-mini)": 79.8,
|
62 |
+
"MLAB (gpt-4o)": 79.8,
|
63 |
+
"MLAB (llama3-1-405b-instruct)": 79.8
|
64 |
},
|
65 |
"backdoor-trigger-recovery": {
|
66 |
"CoI-Agent (o1) + MLAB (gpt-4o)": 85.0,
|