improvements_still_to_do.txt

1st, here's a list of things that just cannot get any better within the
framework of LHC-FASER:

- electroweak contributions to sparticle production
- neutrino contributions to MET
- showering from extra jets from cascades such as those from W bosons


here's a list of things that *might* be feasible to implement, with a hell of a
lot of work & presumably a huge speed hit:

- correlations of lepton acceptance numbers with jet+MET cuts
  (currently lepton cuts are based on all events, rather than just the subset
  that pass jet+MET cuts; to change this, we'd need lepton acceptance grids per
  signal as well)
- correlations of acceptances of leptons from the same cascade
  (currently each lepton is assumed to pass or fail cuts pretty independently
  of the rest of the leptons; to change this, we'd need a lot more work on the
  analytic formulae & probably new lepton grids, & all for probably not much
  improvement in accuracy at all)


here's a list of things that we really should implement, since they should be
straightforward, though maybe time-consuming:

- proper flavor-by-flavor (LO) cross-section grids
  (not sure how trustworthy Prospino is for this though)
- more accurate K-factor grids
  (NNLO or NLL or whatever)
- decoupling of transverse & longitudinal modes of electroweakino cascade
  decays through on-shell vector bosons
  (should be straightforward breakdown in terms of squared couplings, but
  currently the longitudinal parts are assumed to be negligible because of the
  m_fermion / m_Z suppression - BOL is not sure if this is ever compensated by
  small enough Higgsino components suppressing the transverse contribution)
  

here's a list of things that are quite necessary, but are just placeholder
because we're trying to just get a half-working version as soon as we can:

- proper 3-body electroweakino decays
  (at the very least in some limit of just off-shell Z or just off-shell
  selectron, maybe with 1/m^2-type propagators)