Goal (apples-to-apples with MedQA/NIH): Does BARE system authority transfer in imaging? Replace the two peers with a single wrong 'automated pre-screen' flag on the same cued images and compare its contagion to peer-assertion.
Copy this template: experiments/imaging/imaging_system_flag.py. It already implements the experiment; you mostly swap in the MIMIC-CXR image manifest and adapt the loader if needed. Do NOT rewrite the method, replicate it.
What to do: Report system-flag contagion vs peer-assertion contagion (Fisher exact).
Golden rule: real data + a real Gemini API run only. A passing mock / synthetic smoke test does NOT count. Commit results under experiments/mimic_cxr_image/results/ in the SAME JSON/JSONL format as the source, so the cross-dataset comparison table can be filled directly.
Definition of done: the real keyed-run numbers are committed under results/, and the row/cell for MIMIC-CXR image in the cross-dataset table can be read off them. Part of #295.
Goal (apples-to-apples with MedQA/NIH): Does BARE system authority transfer in imaging? Replace the two peers with a single wrong 'automated pre-screen' flag on the same cued images and compare its contagion to peer-assertion.
Copy this template:
experiments/imaging/imaging_system_flag.py. It already implements the experiment; you mostly swap in the MIMIC-CXR image manifest and adapt the loader if needed. Do NOT rewrite the method, replicate it.What to do: Report system-flag contagion vs peer-assertion contagion (Fisher exact).
Golden rule: real data + a real Gemini API run only. A passing mock / synthetic smoke test does NOT count. Commit results under
experiments/mimic_cxr_image/results/in the SAME JSON/JSONL format as the source, so the cross-dataset comparison table can be filled directly.Definition of done: the real keyed-run numbers are committed under results/, and the row/cell for MIMIC-CXR image in the cross-dataset table can be read off them. Part of #295.