Table 2 Step-wise performance of large language model (LLM) annotation metrics. Accuracy for binary classifications in Step 2 and Step 4, assessing correct identifications of node types and regulation types, simplified by specific types
From: Survey on large language model annotation of cellular senescence from figures in review articles
Step | Accuracy |
|---|---|
Step 2: Node-type classification | 0.84 |
Process | 0.83 |
Molecule | 0.84 |
Compound | 0.57 |
Cell | 1.0 |
Disease | 0.98 |
Step 4: positive/negative identification | 0.82 |
Positive regulation | 0.85 |
Negative regulation | 0.66 |