Today, there are 69 new entries on hep-th: 27 of them are new papers. That's a lot and I will surely omit some interesting ones. Nevertheless, this is my selection:
Hořava-Lifshitz theory violates equivalence principle
Amir Esmaeil Mosaffa tries to write the action for particles moving through the Hořava-Lifshitz spacetime. He finds out that not one mass parameter but two mass parameters "M,m", are needed. As a result, the trajectories will depend on their ratio, "m/M", which contradicts the equivalence principle - although, truth to be said, I didn't quite understand whether you can choose "m" and "M" for each object/particle arbitrarily and how they add up for bound states etc.
This is a good example showing how different principles needed for physical consistency are linked to each other. If you sacrifice the Lorentz symmetry, it's likely that the equivalence principle will fail, too.
D=11 SUGRA in pure spinors
Martin Cederwall writes down the action for D=11 supergravity, with manifest supersymmetry, using Berkovits' pure spinor gadgets that you probably know from 10-dimensional backgrounds of string theory only. With a proper field redefinition, the whole action is polynomial and ends with cubic terms!
Topological gravity in D=7
Lu and Pang write down a pretty interesting D=7 topological theory of gravity with extra terms inspired by the Chern-Simons couplings in D=11 supergravity. The new theory could be related to M-theory in a similar way as the topological string models are related to type II string theories (although it's just my comment: the authors never use the word "string" in the paper). Also, they present some solutions to their equations.
Hartle-Hawking and non-Gaussianities
Hartle, Hawking, and Hertog (HHH) try to determine the general qualitative predictions of the Hartle-Hawking wave function (which they awkwardly call NBWF so that you may continue to use the same term as I just did).
Assuming that there is something like NBWF and that you only choose the parts of the multiverse where we qualitatively observe what we have already observed (uniformity of the visible patch etc.), there are two possibilities:
1. There is generically no eternal inflation in the typical histories, and correspondingly a small amount of inflation of our patch. That's correlated with homogeneity that extends beyond the cosmic horizon and with expected large non-Gaussianities.
2. The histories dominated by eternal inflation predict a lot of inflation in our Universe, significant non-uniformities behind the cosmic horizon, and almost no non-Gaussianities.
So if this NBWF-based analysis is right and if you will be able to look behind the horizon ;-), you will be able to predict something about the non-Gaussianities of the cosmic microwave background - and vice versa: if you know the answer to non-Gaussianities, you will be able to predict something about the beyond-the-horizon physics that no one can see. ;-)
That's helpful but in some sense, I really view physics behind the cosmic horizon to be unphysical - because of the complementarity as well as the normal causality. There are many other question marks.
For example, the scenarios (1) and (2) are so qualitatively different that I believe that only one of them is relevant in real physics - we just don't know which one. For this reason, it looks somewhat redundant to try to cover both of them in the same framework.