OpenDecoder: Open Large Language Model Decoding to Incorporate Document Quality in RAG
This addresses the issue of noisy context in RAG systems for users relying on LLM-generated answers, though it is incremental as it builds on existing RAG methods by adding quality evaluation features.
The paper tackles the problem of variable relevance and usefulness of retrieved documents in retrieval-augmented generation (RAG) by proposing OpenDecoder, which incorporates explicit quality indicators like relevance, ranking, and QPP scores into generation; it demonstrates effectiveness and better robustness by outperforming baselines on five benchmark datasets.
The development of large language models (LLMs) has achieved superior performance in a range of downstream tasks, including LLM-based retrieval-augmented generation (RAG). The quality of generated content heavily relies on the usefulness of the retrieved information and the capacity of LLMs' internal information processing mechanism to incorporate it in answer generation. It is generally assumed that the retrieved information is relevant to the question. However, the retrieved information may have a variable degree of relevance and usefulness, depending on the question and the document collection. It is important to take into account the relevance of the retrieved information in answer generation. In this paper, we propose OpenDecoder, a new approach that leverages explicit evaluation of the retrieved information as quality indicator features for generation. We aim to build a RAG model that is more robust to varying levels of noisy context. Three types of explicit evaluation information are considered: relevance score, ranking score, and QPP (query performance prediction) score. The experimental results on five benchmark datasets demonstrate the effectiveness and better robustness of OpenDecoder by outperforming various baseline methods. Importantly, this paradigm is flexible to be integrated with the post-training of LLMs for any purposes and incorporated with any type of external indicators.