Establish survey-scale redshift recovery and deblending performance

Establish the purity, completeness, catastrophic-failure rate, line-flux bias, and computational scalability of the three-orientation image-sliced forward-model extraction for the proposed 3.5-meter Segmented-Mirror Robotic Space Telescope survey.

Background

The scientific forecasts assume that the joint sparse inversion can recover reliable redshifts from crowded, multi-orientation slitless observations. However, the figures and preliminary forced-line tests demonstrate geometry rather than blind redshift-recovery performance.

The paper identifies survey-scale validation with realistic injected galaxies, calibrated detector scenes, and the flight extraction method as a required unresolved task. It also specifically identifies the computational cost of processing millions of sources and thousands of visits as unmeasured.

References

Survey-scale tests have not yet established the required purity, completeness, or flux-bias limits. Figure~\ref{fig:crowded} illustrates the deblending geometry on one tile and Figure~\ref{fig:slicer} illustrates the adopted image-slicer mapping. Neither figure supplies recovery statistics. The computational cost of the joint sparse inversion also remains unmeasured for approximately $106$--$3\times106$ sources across $104$ orientation visits.

— 3.5-meter Segmented-Mirror Robotic Space Telescope Mission White Paper II. Key Scientific Mission: Wide-Field Cosmology and Galaxy Evolution  (2609.02574 - Kim et al., 2 Sep 2026) in Section 2.8, “Validation with Simulated Detector Images”