Uncertainty Quantification for Retrieval-Augmented Reasoning

Retrieval-augmented reasoning (RAR) is a recent evolution of retrieval-augmented generation (RAG) that employs multiple reasoning steps for retrieval and generation. While effective for some complex queries, RAR remains vulnerable to errors and misleading outputs. Uncertainty quantification (UQ) offers methods to estimate the confidence of systems' outputs. These methods, however, often handle simple queries with no retrieval or single-step retrieval, without properly handling RAR setup. Accurate estimation of UQ for RAR requires accounting for all sources of uncertainty, including those arising from retrieval and generation. In this paper, we account for these sources and introduce Retrieval-Augmented Reasoning Consistency (R 2 C), a novel UQ method for RAR. The core idea of R 2 C is to perturb the multi-step reasoning process by applying various actions to reasoning steps. These perturbations alter the retriever's input, which shifts its output and consequently modifies the generator's input at the next step. Through this iterative feedback loop, the retriever and generator continuously reshape each other's inputs, enabling us to capture uncertainty arising from both components. Experiments on five popular RAR systems across diverse QA datasets show that R 2 C improves AU-ROC by over 5% on average compared to the state-of-the-art UQ baselines. Extrinsic evaluations using R 2 C as an external signal further confirm its effectiveness for two downstream tasks: in the Abstention task, it achieves 5% gains in both F1Abstain and Ac-cAbstain; in Model Selection, it improves exact match by 7% over single models and 3% over selection methods. Code is available on https://github.com/HeydarSoudani/R2C.