Left panel: The distributions of distances between the mean and the normalised mean ℳ_* for the retrieved neighbours and the target object, i.e. ${\mathcal{M}}_{\ast }^{\mathrm{t}\mathrm{a}\mathrm{r}\mathrm{g}}-⟨{\mathcal{M}}_{\ast }^{\mathrm{n}\mathrm{e}\mathrm{i}\mathrm{g}}⟩$ (by blue colour) and ${\mathcal{M}}_{\ast }^{\mathrm{t}\mathrm{a}\mathrm{r}\mathrm{g}}-{\mathcal{M}}_{\ast }^{\mathrm{n}\mathrm{o}\mathrm{r}\mathrm{m}}$ (by green colour), for all, star-forming and quiescent subsamples of target objects. The black dashed line represents the cut at 1 dex/$\sqrt{2}$ used to define ULISSE efficiency (see description in Section 4.3). Right panel: The same as on the left panel, but for SFR^targ - ⟨SFR^neig⟩ and SFR^targ - SFR^norm.