Problem
Research questions and friction points this paper is trying to address.
Agentic RAG
Static RAG
Taiwanese Historical Question Answering
Complementarity
Performance Comparison
Bridging Static and Agentic RAG for Taiwanese Historical Question Answering
๐ค AI Summary
็ ็ฉถ้่ฟๅฏนๆฏ้ๆๅๅจๆRAGๅจๅฐๆนพๅๅฒ้ฎ็ญไธญ็่กจ็ฐ๏ผๆๅบไธ็งๅๅค็้ๆฉๅจไปฅ็ปๅไธค่
ไผๅฟ๏ผๆ้ซๅ็ญๅ็กฎๆงใ
๐ Abstract
Agentic retrieval-augmented generation (RAG) enables language models to adapt retrieval based on previously retrieved evidence, but it remains unclear whether such adaptive orchestration consistently outperforms well-designed static pipelines. We conduct a controlled comparison of agentic and static RAG for Taiwanese historical question answering, sharing the same generator and hybrid retrieval backend. Despite similar aggregate performance, the two pipelines differ on 70.83% of questions, with their advantages largely canceling out when averaged. An oracle that selects the better response per question improves the composite score by 0.2417 over the better individual pipeline, revealing substantial headroom for question-level selection. We therefore introduce a post-hoc selector that compares the two responses and their cited evidence, significantly outperforming either individual pipeline and recovering 60.34% of the oracle headroom. These results show that aggregate comparisons can obscure meaningful question-level differences between retrieval strategies, suggesting that exploiting their complementarity may be more fruitful than seeking a universally superior pipeline.
Innovation
Methods, ideas, or system contributions that make the work stand out.
agentic RAG
static RAG
post-hoc selector
complementarity
๐ Similar Papers
No similar papers found.
