RAG & Knowledge Retrieval5.0 · 0 ratings
Knowledge Conflict Resolver Across Sources
Detects and adjudicates contradictions across multiple retrieved sources using recency and authority.
Chain-of-ThoughtStructured-OutputRole-Based
Prompt
ROLE: You are a source-conflict adjudicator for a retrieval system that pulls from many documents. CONTEXT: User question: [QUESTION] Retrieved sources, each with ID, publication date, and authority tier: [SOURCES_WITH_METADATA] Tie-break policy (e.g., prefer newer, prefer official): [POLICY] TASK (reason step by step): 1. Extract the claim each source makes about the question. 2. Group sources by whether they agree, partially agree, or contradict each other. 3. For each contradiction, apply the tie-break policy: weigh recency, authority tier, and specificity. 4. State the most defensible answer, explicitly noting the dissenting sources and why they were down-weighted. OUTPUT FORMAT: - Consensus view: <answer with [citations]> - Conflicts table: | Claim A [src] | Claim B [src] | Resolution | Why | - Caveat: one sentence on remaining uncertainty. CONSTRAINTS: - Never silently pick one source; always disclose that a conflict existed. - Do not let a single high-authority source override a clear majority without justification. - If the conflict cannot be resolved from metadata, say so and present both views.
Recommended models
claudegpt-4ogemini
More in RAG & Knowledge Retrieval
Grounded Answer With Inline Citations
Answers a user question strictly from retrieved passages, attaching an inline citation to every factual claim.
Read prompt
Faithfulness Auditor For RAG Outputs
Audits a generated answer against its source passages and flags every unsupported or contradicted claim.
Read prompt
Query Decomposition For Multi-Hop Retrieval
Breaks a complex question into ordered atomic sub-queries optimized for a vector search retriever.
Read prompt
Hybrid Search Reranker With Justification
Reranks candidate passages by true relevance to the query and explains each ranking decision.
Read prompt