Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Forward Prediction on Materials Discovery (In-Domain)
Loading...
51.2
Overall Score
ARIA-FULL
21.976
29.563
37.15
44.737
Jun 21, 2026
Overall Score
Scientific Accuracy
Functional Equivalence
Completeness
Interpretability
Updated 1mo ago
Evaluation Results
Method
Method
Links
Overall Score
Scientific Accuracy
Functional Equivalence
Completeness
Interpretability
ARIA-FULL
2026.06
51.2
25.4
33.9
93
92.5
ARIA-FULL
2026.06
51.2
-
-
-
-
ARIA-SEARCH
2026.06
51.1
27.5
38.2
79.9
81.6
Self-RAG
2026.06
45.3
-
-
-
-
ARIA-CORE
2026.06
41
23.5
40
74.9
79.1
Baseline LLM
2026.06
34
25.2
38.5
65.9
87.7
NAIVE-KG
2026.06
33.7
28
39.4
73.2
35
KG-ONLY
2026.06
23.1
6.8
14.3
60
35
Feedback
Search any
task
Search any
task