Record Grouping Controls Evidence Weight in Language Models

Repeated retrieval records can make one evidence source appear to provide several independent contributions. We group records before generation, remove exact within-group copies, and preserve complementary content in one record per supplied group. In the central natural-text experiment, replacing four falsely split records with one full-content group reduces the normalized likelihood share of the attacker-favored candidate by 18.2–40.7% across all eight model–dataset pairs, while retaining the same 160 evidence words. A stricter six-slot control holds headers, object count, and token length fixed and yields reductions of 1.2–15.6%. On 120 real questions with overlapping distractor excerpts, estimated document grouping improves answer accuracy by 5.42 and 5.83 percentage points for Mistral-7B and Phi-4.