decision-models-for-industrial-image-qc

Progress

Milestones

Milestones
IDMilestoneStatePlanTasks complete
M1Tune JEV on validation, test once, compare with CLIP and Clef-Flash (MSEC 2027 abstract)done--
M2Full MSEC 2027 paper: fmm and yjc write the text and make the method figure (D93, D106). Claude Code makes the data figures and the LaTeX tables in the style of fmm (D96, D97, D107) and gives each reference in the .bib file a doi field (D93). fmm did the literature search for the introduction and the framework, so Claude Code does no literature search (D260). Following the ASME position statement on AI, Claude Code writes no text of the paper and no observations (D105), and the paper discloses how AI tools were used. The paper is in the private repository fmegahed/decision-models-paper, linked to Overleaf (D94)now--
M3R package decisions (named decisionqc until D146; design D183, plan D187)doneplans/2026-10-23-decisions-package-plan.md12 of 13
M4Record additions: releases, results, timeline and contributions (replication-materials 1.2.0)doneplans/2026-10-20-record-additions-plan.md8 of 8
M5Fair comparison: shared label budget, blind description loop, calibration of all methods (v0.0.2)doneplans/2026-10-20-fair-comparison-plan.md11 of 12
M6Hugging Face Space of decisions: a Shiny app in Docker that asks Jev a question about each image or text, with the R code and a bundle of the run (design D242, plan D243)nowplans/2026-10-24-decisions-space-plan.md-
M7Box study, then Box in decisions 0.2.0 and the Space: find defects with the TypeSafe Box primitive (Jev 1.14) on FabriD and the AutoVI classes tank screw and pipe clip, with a search over regions of several sizes that overlap, as in imageST (D268 to D274)nowplans/2026-10-28-box-agentic-plan.md-

Open items

Open items and blockers
IDItemBlocksOwner
O1On hold (D104). Test the refinement step (criteria and references) with smaller validation sets; test balanced accuracy settled at 10 to 20 validation images per class, but the ECE kept falling up to 50 (figure F7 of v0.0.1).-fmm
O3On hold (D104). Extrusion frames come from video: consider a time-based split, because near-identical frames help methods with large learning sets.-fmm
O5State in the paper that the audit of the description loop finds plausible reads outside the sandbox, not deliberate evasion: an interpreter one-liner (python -c, Rscript -e) can name a parent folder without a slash or '..', and the audit does not see it (ledger of M5, Task 4).M2fmm
O6Write the AI-use disclosure that ASME asks for (position statement, section B): tools, versions, maker, access, dates and uses in the Acknowledgements, and how AI was used in the Methods. Claude Code collects the facts from the record; fmm and yjc write the text.M2fmm

Paid calls

Paid calls by service
ServiceCallsMax callsCostMax cost
elevenlabs332000.368 USD2 USD
openai552487001.483 USD20.5 USD
openrouter1495500.001 USD10 USD
typesafe14602704462.64 USD44.65 USD
typesafe-space0577000 USD20 USD