BlendFilter: Advancing Retrieval-Augmented Large Language Models via Query Generation Blending and Knowledge Filtering
This work addresses challenges in retrieval-augmented LLMs for knowledge-intensive tasks, offering a novel method that is incremental in enhancing existing approaches.
The paper tackles the problem of noisy knowledge retrieval in retrieval-augmented large language models for complex inputs, introducing BlendFilter, which integrates query generation blending and knowledge filtering to achieve significant performance improvements over state-of-the-art baselines on three open-domain question answering benchmarks.
Retrieval-augmented Large Language Models (LLMs) offer substantial benefits in enhancing performance across knowledge-intensive scenarios. However, these methods often face challenges with complex inputs and encounter difficulties due to noisy knowledge retrieval, notably hindering model effectiveness. To address this issue, we introduce BlendFilter, a novel approach that elevates retrieval-augmented LLMs by integrating query generation blending with knowledge filtering. BlendFilter proposes the blending process through its query generation method, which integrates both external and internal knowledge augmentation with the original query, ensuring comprehensive information gathering. Additionally, our distinctive knowledge filtering module capitalizes on the intrinsic capabilities of the LLM, effectively eliminating extraneous data. We conduct extensive experiments on three open-domain question answering benchmarks, and the findings clearly indicate that our innovative BlendFilter surpasses state-of-the-art baselines significantly.