Feedback Adaptation for Retrieval-Augmented Generation
This addresses a previously overlooked dimension of RAG system behavior in interactive settings, though it is incremental as it focuses on evaluation and a specific method rather than a new paradigm.
The paper tackles the problem of evaluating how Retrieval-Augmented Generation (RAG) systems adapt to corrective feedback, showing that training-based approaches have a trade-off between delayed correction and reliable adaptation, while their proposed PatchRAG method achieves immediate correction and strong generalization without retraining.
Retrieval-Augmented Generation (RAG) systems are typically evaluated under static assumptions, despite being frequently corrected through user or expert feedback in deployment. Existing evaluation protocols focus on overall accuracy and fail to capture how systems adapt after feedback is introduced. We introduce feedback adaptation as a problem setting for RAG systems, which asks how effectively and how quickly corrective feedback propagates to future queries. To make this behavior measurable, we propose two evaluation axes: correction lag, which captures the delay between feedback provision and behavioral change, and post-feedback performance, which measures reliability on semantically related queries after feedback. Using these metrics, we show that training-based approaches exhibit a trade-off between delayed correction and reliable adaptation. We further propose PatchRAG, a minimal inference-time instantiation that incorporates feedback without retraining, demonstrating immediate correction and strong post-feedback generalization under the proposed evaluation. Our results highlight feedback adaptation as a previously overlooked dimension of RAG system behavior in interactive settings.