Some hours ago, the ATLAS collaboration posted two papers on its website:
Evidence for Higgs boson Yukawa couplings in the \(H \to \tau\tau\) decay mode with the ATLAS detectorSpoilers alert. Too late. The result is that all the basic figures are found to be in almost exact agreement with the Standard Model.
Observation and measurement of Higgs boson decays to \(WW^*\) with ATLAS at the LHC
The first paper involving the \(\tau\)-leptons may be interesting because it's the same decay channel for which ATLAS organized the Kaggle contest. Recall that your humble correspondent dropped from 1st place to 8th place when the "private dataset" was used at the end.
Also CERN-related: a new CERN 5-minute video about the rise of the Standard Model.
The winner, Gábor Melis, was the only one who has used the neural networks to solve the challenge. Everyone else at the top, including myself, used algorithms based on the boosted decision trees (e.g. xgboost), and this paradigm was exploited in the newly revealed paper, too. However, there are almost no details so we can't know whether the authors of the paper would have a chance to compete with us in the contest. At the end, it doesn't matter. The significance level itself has some statistical fluctuation in it, plus minus one sigma, so whether a Hungarian guy had a score better than me by 0.04 really doesn't matter in physics.
The second paper is about the \(WW^*\) decays of the Higgs boson. It's a pair of two \(W\)-bosons but, as the asterisk indicates, one of them is virtual. Again, ATLAS finds the cross section to agree with the Standard Model extremely well. I feel that it could be bad news for the hint of new physics in the \(WW\)-decays of the Higgs.
Previously, ATLAS saw an enhanced cross section if both \(W\)-bosons are on-shell. If one of them is virtual, the excess seems to go away. It may look a bit strange and may be an indication that the excess in the on-shell \(WW\)-decays was a fluke. But maybe there is a reason not to expect an excess in the \(WW^*\) analysis...