Large quasar groups are the sky's most extravagant coincidences: collections of dozens of quasars spanning 300 to 600 million light-years and beyond, with the record-holding Huge-LQG stretching roughly 4 billion light-years and containing 73 quasars at redshift near 1.3 (Clowes et al. 2013; Lietzen et al. 2011). Quasars are rare, briefly active objects, which makes finding them congregated across such volumes statistically expensive.
The expense lands on ΛCDM twice. First, the clustering: quasars should trace the underlying matter field with roughly galaxy-level clustering strength, and the probability of assembling coherent groups at these scales from standard clustering is very low. Second, the principle: the Huge-LQG's extent exceeds the conventional homogeneity scale by several times, joining the great walls and giant arcs in the census of structures that should not exist if the universe averages out at a few hundred megaparsecs. The standard responses question the groups' reality, since friends-of-friends algorithms on sparse catalogs can link unrelated objects, an argument with genuine force that has kept the LQG census contested without ever quite dissolving it.
The standing is tied to the giant-structure family's collective weight: each member is arguable, the population is harder to dismiss. LSST and Euclid quasar catalogs will multiply the sample enormously, testing whether rare-object coherence at gigaparsec scales is an algorithmic mirage or a property of the universe.