Fetching from the wire…
01
02
03
04
05
06
07
08
Research2026-08-21 · source-backed
A synthetic benchmark constructs conflicts where exactly one evidence source matches ground truth, independently varying modality, recency, stated reliability, and provenance. Across open-weight instruction-tuned models the arbitration is systematic: distinct text-versus-number preferences, temporal recency weighted more consistently than explicit reliability cues, and over-trust of external forecasts contradicting direct in-context evidence. arXiv An out-of-date but recent-looking tool response quietly outranks a correct fact you put in the prompt. Labeling your sources "authoritative" does less than you think.
Each link below shares sources, entities, or timing with this story.
Shared entity: When / Same source domain / Shared topic / Earlier coverage / Tension
Both cover When; reported by the same outlet (arxiv.org); overlapping topics (correct, model, tool).
Both cover When; reported by the same outlet (arxiv.org); overlapping topics (model, tool).
Shared entity: When / Same source domain / Earlier coverage / Tension / Downstream implication
Both cover When; reported by the same outlet (arxiv.org); earlier When coverage from 2026-08-10.
Shared entity: When / Same source domain / Shared topic / Earlier coverage
Both cover When; reported by the same outlet (arxiv.org); overlapping topics (evidence, reliability).
Both cover When; reported by the same outlet (arxiv.org); overlapping topics (conflict, consistently).
Both cover When; reported by the same outlet (arxiv.org); overlapping topics (correct, evidence).
Both cover When; reported by the same outlet (arxiv.org); overlapping topics (model, prompt).
Shared entity: When / Same source domain / Earlier coverage / Downstream implication
Both cover When; reported by the same outlet (arxiv.org); earlier When coverage from 2026-08-19.