Tag
This paper introduces a taxonomy and dataset for contextual knowledge conflicts in large language models, experiments with seven LLMs, and proposes a steering method to improve conflict resolution in reasoning and summarization tasks.
ConflictRAG is a conflict-aware RAG framework that detects, classifies, and resolves knowledge conflicts in retrieved documents, achieving 88.7% detection F1 and 5.3–6.1% correctness gains over baselines while reducing API costs by 62%.
This paper investigates how LLMs handle knowledge conflicts in retrieval-augmented generation by studying their preferences for different information sources. The authors find that LLMs prefer institutionally-corroborated sources but these preferences can be reversed by repetition, proposing a method to reduce repetition bias while maintaining consistent source preferences.