Knowledge-to-Action QA on 50 Knowledge-to-Action Questions (gold-standard)
29.9ROUGE-1vanilla LLM
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| vanilla LLMBackbone LLM=mistral-nemo-it2025.10 | 29.9 | 27.1 | |
| text-chunk RAGBackbone LLM=gemma-3-it2025.10 | 29.6 | 27.6 | |
| vanilla LLMBackbone LLM=gemma-3-it2025.10 | 29.3 | 27.3 | |
| KEOBackbone LLM=gemma-3-it2025.10 | 29.3 | 27.2 | |
| KEOBackbone LLM=mistral-nemo-it2025.10 | 29.2 | 26.8 | |
| text-chunk RAGBackbone LLM=mistral-nemo-it2025.10 | 21.3 | 19.2 | |
| KEOBackbone LLM=phi-42025.10 | 12.4 | 10.6 | |
| vanilla LLMBackbone LLM=phi-42025.10 | 11 | 10 | |
| text-chunk RAGBackbone LLM=phi-42025.10 | 10.7 | 9.3 |