Do LLMs Make More Mistakes If They Do Not Believe the Input Data?

arXiv:2609.09363v1 Announce Type: new Abstract: Large language models (LLMs) are prone to hallucinating or misinterpreting facts, which impairs their usability in retrieval-augmented generation or data-to-text systems. We analyse how faithfulness of LLMs to provided context depends on how plausible they perceive the context to be (context-memory conflict). To better identify error patterns, we make use of the increased difficulty of non-English and low-resource language text generation and input data based on local knowledge, only partially captured in models' parametric knowledge. We let the models generate text in English, Czech, Slovak and Upper Sorbian from factual (FA), counterfactual (CFA) and fictional (FI) RDF triples containing local Czech and Slovak data. Contrary to our expectations, we observe only a weak context-memory conflict on the human-annotated sample. For Kimi K3 as an LLM judge, which agrees well with human annotations on the sample, counterfactual inputs receive only slightly lower faithfulness scores than factual ones (-0.05 on a 1-5 scale). We also find that a suboptimal choice of LLM judge would lead to overestimating the strength of the context-memory conflict.

Related

• Retrieval-Augmented Large Language Models as Components of Cognitive Computing architecture for Regulatory Knowledge Management
• TagRAG: Tag-guided Hierarchical Knowledge Graph Retrieval-Augmented Generation
• Leveraging External Knowledge for Historical Document Restoration via Retrieval-Augmented Large Language Models
• Curated retrieval versus open web search in public AI information services: a coverage-trust trade-off
• Self-Evaluation Is Already There: Eliciting Latent Judge Calibration in Base LLMs with Minimal Data

Source: arXiv cs.CL | 2026-09-10