Question Answering on ARC Challenge (Perf. and Sp.)
83.1AccuracyBest Model
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Best ModelBackbone=LLaMA 3.1 8B, Shots=5-shot, #D=32025.10 | 83.1 | 1.17 | |
| BSBABackbone=LLaMA 3.1 8B, Shots=5-shot, #D=92025.10 | 78.62 | 1.42 | |
| BaselineBackbone=LLaMA 3.1 8B, Shots=5-shot2025.10 | 78.2 | — |