PrismRAG: Boosting RAG Factuality with Distractor Resilience and Strategized Reasoning
It addresses factuality issues in RAG for question-answering applications, representing an incremental improvement over existing methods.
The paper tackles the problem of retrieval-augmented generation (RAG) falling short due to confusing semi-relevant passages and lack of deep reasoning, proposing PrismRAG, which improves average factuality by 5.4% across 12 benchmarks.
Retrieval-augmented generation (RAG) often falls short when retrieved context includes confusing semi-relevant passages, or when answering questions require deep contextual understanding and reasoning. We propose an efficient fine-tuning framework, called PrismRAG, that (i) trains the model with distractor-aware QA pairs mixing gold evidence with subtle distractor passages, and (ii) instills reasoning-centric habits that make the LLM plan, rationalize, and synthesize without relying on extensive human engineered instructions. Evaluated across 12 open-book RAG QA benchmarks spanning diverse application domains and scenarios, PrismRAG improves average factuality by 5.4%, outperforming state-of-the-art solutions.