Safety
ERA: Evidence-based Reliability Alignment for Honest Retrieval-Augmented Generation
arXiv:2604.20854v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) grounds language models in factual evidence but introduces critical challenges regarding knowledge conflicts betw
arXiv:2604.20854v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) grounds language models in factual evidence but introduces critical challenges regarding knowledge conflicts between internalized parameters and retrieved information. However, existing reliability methods, typically relying on scalar confidence, fail to explicitly distinguish between epistemic uncertainty and inherent data ambiguity in such hybrid scenarios. In this paper, we propose a new framework called ERA (Evidence-based Reliability Alignment) to enhance abstention behavior in RAG systems by shifting confidence estimation from scalar probabilities to explicit evidence distributions. Our method consists of two main components: (1) Contextual Evidence Quantification, which models internal and external knowledge as independent belief masses via the Dirichlet distribution, and (2) Quantifying Knowledge Conflict, which leverages Dempster-Shafer Theory (DST) to rigorously measure the geometric discordance between information sources. These components are used to disentangle epistemic uncertainty from aleatoric uncertainty and modulate the optimization objective based on detected conflicts. Experiments on standard benchmarks and a curated generalization dataset demonstrate that our approach significantly outperforms baselines, optimizing the trade-off between answer coverage and abstention with superior calibration.
Related
- GRAIL:Learning to Interact with Large Knowledge Graphs for Retrieval Augmented Reasoning
- ORPHEAS: A Cross-Lingual Greek-English Embedding Model for Retrieval-Augmented Generation
- RELOOP: Recursive Retrieval with Multi-Hop Reasoner and Planners for Heterogeneous QA
- Beyond Factual Grounding: The Case for Opinion-Aware Retrieval-Augmented Generation
- AtomicRAG: Atom-Entity Graphs for Retrieval-Augmented Generation
- Coverage, Not Averages: Semantic Stratification for Trustworthy Retrieval Evaluation
Source: arXiv cs.AI | 2026-04-24