KP-Comment Factual Alignment on AMAZONKP (test)
0.749AlignScoreQQSUM-RAG
Evaluation Results
| Method | Links | |
|---|---|---|
| QQSUM-RAGBase Framework=QQSUM-RAG, LLM/Model Variant=Mistral2025.06 | 0.749 | |
| PAKPABase Framework=Frozen Retriever + KPA, LLM/Model Variant=PAKPA2025.06 | 0.749 | |
| Frozen Retriever + prompt LLMBase Framework=Frozen Retriever + prompt LLM, LLM/Model Variant=GPT-4-Turbo2025.06 | 0.673 | |
| (Retriever + LLM)co-trainBase Framework=(Retriever + LLM)co-train, LLM/Model Variant=Mistral2025.06 | 0.653 | |
| QQSUM-RAGBase Framework=QQSUM-RAG, LLM/Model Variant=Vicuna2025.06 | 0.63 | |
| Frozen Retriever + prompt LLMBase Framework=Frozen Retriever + prompt LLM, LLM/Model Variant=Mistral2025.06 | 0.624 | |
| Frozen Retriever + prompt LLMBase Framework=Frozen Retriever + prompt LLM, LLM/Model Variant=Vicuna2025.06 | 0.531 | |
| (Retriever + LLM)co-trainBase Framework=(Retriever + LLM)co-train, LLM/Model Variant=Vicuna2025.06 | 0.394 | |
| RKPA-BaseBase Framework=Frozen Retriever + KPA, LLM/Model Variant=RKPA-Base2025.06 | 0.354 |