Graph Reasoning on NLGraph
97Success RateEGL-SCA (Full)
Evaluation Results
| Method | Links | |
|---|---|---|
| EGL-SCA (Full)Base LLM=gpt-5.4-nano2026.05 | 97 | |
| w/o Protocol-Reliability ObjBase LLM=gpt-5.4-nano, Ablation=w/o Protocol-Reliability Obj2026.05 | 94.5 | |
| w/o SCA RoutingBase LLM=gpt-5.4-nano, Ablation=w/o SCA Routing2026.05 | 93 | |
| w/o Tool GrowthBase LLM=gpt-5.4-nano, Ablation=w/o Tool Growth2026.05 | 93 | |
| w/o Instruction Evol.Base LLM=gpt-5.4-nano, Ablation=w/o Instruction Evol.2026.05 | 88.5 | |
| AgentSquareBase LLM=gpt-5.4-nano, Category=Tool-Augmented Agents, Tool Access=Fixed Toolbox2026.05 | 63.5 | |
| MA-GTSBase LLM=gpt-5.4-nano, Category=Tool-Augmented Agents, Tool Access=Fixed Toolbox2026.05 | 62.9 | |
| ReActBase LLM=gpt-5.4-nano, Category=Tool-Augmented Agents, Tool Access=Fixed Toolbox2026.05 | 59 | |
| FixedSolverBase LLM=gpt-5.4-nano, Category=Tool-Augmented Agents, Tool Access=Fixed Toolbox2026.05 | 58 | |
| w/o Tool UseBase LLM=gpt-5.4-nano, Ablation=w/o Tool Use2026.05 | 44.5 | |
| Chain-of-Thought (CoT)Base LLM=gpt-5.4-nano, Category=Prompting & Self-Improvement, Tool Access=No Tools2026.05 | 30 | |
| Direct PromptingBase LLM=gpt-5.4-nano, Category=Prompting & Self-Improvement, Tool Access=No Tools2026.05 | 24.5 | |
| GEPABase LLM=gpt-5.4-nano, Category=Prompting & Self-Improvement, Tool Access=No Tools2026.05 | 21.5 | |
| ExpeLBase LLM=gpt-5.4-nano, Category=Prompting & Self-Improvement, Tool Access=No Tools2026.05 | 21 | |
| ReflexionBase LLM=gpt-5.4-nano, Category=Prompting & Self-Improvement, Tool Access=No Tools2026.05 | 17 | |
| Few-ShotBase LLM=gpt-5.4-nano, Category=Prompting & Self-Improvement, Tool Access=No Tools2026.05 | 15 |