Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Graph Reasoning on NLGraph
Loading...
97
Success Rate
EGL-SCA (Full)
11.72
33.86
56
78.14
May 11, 2026
Success Rate
Updated 22d ago
Evaluation Results
Method
Method
Links
Success Rate
EGL-SCA (Full)
Base LLM=gpt-5.4-nano
2026.05
97
w/o Protocol-Reliability Obj
Base LLM=gpt-5.4-nano,...
2026.05
94.5
w/o SCA Routing
Base LLM=gpt-5.4-nano,...
2026.05
93
w/o Tool Growth
Base LLM=gpt-5.4-nano,...
2026.05
93
w/o Instruction Evol.
Base LLM=gpt-5.4-nano,...
2026.05
88.5
AgentSquare
Base LLM=gpt-5.4-nano,...
2026.05
63.5
MA-GTS
Base LLM=gpt-5.4-nano,...
2026.05
62.9
ReAct
Base LLM=gpt-5.4-nano,...
2026.05
59
FixedSolver
Base LLM=gpt-5.4-nano,...
2026.05
58
w/o Tool Use
Base LLM=gpt-5.4-nano,...
2026.05
44.5
Chain-of-Thought (CoT)
Base LLM=gpt-5.4-nano,...
2026.05
30
Direct Prompting
Base LLM=gpt-5.4-nano,...
2026.05
24.5
GEPA
Base LLM=gpt-5.4-nano,...
2026.05
21.5
ExpeL
Base LLM=gpt-5.4-nano,...
2026.05
21
Reflexion
Base LLM=gpt-5.4-nano,...
2026.05
17
Few-Shot
Base LLM=gpt-5.4-nano,...
2026.05
15
Feedback
Search any
task
Search any
task