Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Abductive Logical Reasoning on QUAIL
Loading...
84
Accuracy (QUAIL)
PACS Llama 3.3 70B
54.88
62.44
70
77.56
May 8, 2026
Accuracy (QUAIL)
Updated 23d ago
Evaluation Results
Method
Method
Links
Accuracy (QUAIL)
PACS Llama 3.3 70B
Method=PACS, Model=Lla...
2026.05
84
PACS
Backbone=Llama 3.3 70B
2026.05
84
If-Beam
Backbone=Llama 3.3 70B
2026.05
82
SC OSS-GPT-120B
Method=SC, Model=OSS-G...
2026.05
80
PACS Llama 3 8B
Method=PACS, Model=Lla...
2026.05
80
PACS
Backbone=Llama 3-Instr...
2026.05
80
ARGOS
Backbone=Llama 3.3 70B
2026.05
80
SC-20
Backbone=Llama 3.3 70B
2026.05
75
If-Beam
Backbone=Llama 3-Instr...
2026.05
74
ARGOS
Backbone=Llama 3-Instr...
2026.05
73
LoT
Backbone=Llama 3.3 70B
2026.05
72
COT
Backbone=Llama 3.3 70B
2026.05
72
SC-20
Backbone=Llama 3-Instr...
2026.05
70
COT
Backbone=Llama 3-Instr...
2026.05
66
LoT
Backbone=Llama 3-Instr...
2026.05
56
LLM-Tree
Backbone=Llama 3-Instr...
2026.05
56
LLM-Tree
Backbone=Llama 3.3 70B
2026.05
56
Feedback
Search any
task
Search any
task
