Search papers, labs, and topics across Lattice.
ArbGraph is introduced as a pre-generation framework for long-form RAG that resolves factual conflicts by decomposing retrieved documents into atomic claims and organizing them into a conflict-aware evidence graph. An intensity-driven iterative arbitration mechanism propagates credibility signals through the graph to suppress unreliable claims before generation. Experiments on LongFact and RAGChecker demonstrate that ArbGraph improves factual recall and information density while reducing hallucinations and sensitivity to retrieval noise, especially under conflicting evidence.
Untangling evidence validation from text generation, ArbGraph offers a way to build more reliable long-form RAG systems by explicitly resolving factual conflicts *before* generation even begins.
Retrieval-augmented generation (RAG) remains unreliable in long-form settings, where retrieved evidence is noisy or contradictory, making it difficult for RAG pipelines to maintain factual consistency. Existing approaches focus on retrieval expansion or verification during generation, leaving conflict resolution entangled with generation. To address this limitation, we propose ArbGraph, a framework for pre-generation evidence arbitration in long-form RAG that explicitly resolves factual conflicts. ArbGraph decomposes retrieved documents into atomic claims and organizes them into a conflict-aware evidence graph with explicit support and contradiction relations. On top of this graph, we introduce an intensity-driven iterative arbitration mechanism that propagates credibility signals through evidence interactions, enabling the system to suppress unreliable and inconsistent claims before final generation. In this way, ArbGraph separates evidence validation from text generation and provides a coherent evidence foundation for downstream long-form generation. We evaluate ArbGraph on two widely used long-form RAG benchmarks, LongFact and RAGChecker, using multiple large language model backbones. Experimental results show that ArbGraph consistently improves factual recall and information density while reducing hallucinations and sensitivity to retrieval noise. Additional analyses show that these gains are evident under conflicting or ambiguous evidence, highlighting the effectiveness of evidence-level conflict resolution for improving the reliability of long-form RAG. The implementation is publicly available at https://github.com/1212Judy/ArbGraph.
CMU ML