A team tests whether disappearing connections change the rules of quantum entanglement — and finds they mostly don't.
Imagine a model of reality where every particle talks to every other particle, all the time. That's the Sachdev-Ye-Kitaev model, a peculiar quantum system that physicists use to explore black holes and the birth of spacetime itself. Now imagine unplugging most of those conversations—cutting the connections until only a handful remain.
Common sense says: break enough links, and the whole system should collapse into something fundamentally different. The holographic structure linking quantum mechanics to gravity should crumble. But according to new numerical findings, it doesn't. Not quite. Even when researchers severed connections so aggressively that the model retained only a skeleton of its original structure, a specific signature of quantum entanglement—one that flags systems with gravitational duals—persisted almost unchanged.
The result raises a question: What are we actually testing when we test for holography?
When Fewer Ingredients Make the Same Dish
The Sachdev-Ye-Kitaev model wasn't built to be realistic. It's a zero-dimensional quantum system—no space, just particles interacting through random couplings. Yet at low temperatures and large particle counts, this abstract toy behaves like a black hole. Specifically, it mirrors properties of Jackiw-Teitelboim gravity, a simplified two-dimensional gravitational theory.
That correspondence makes it a test bed for holography, the principle that gravity in a higher-dimensional space can emerge from quantum mechanics living on its boundary. In field theories known to have gravity duals, the entanglement entropy between spatial regions obeys strict inequalities. One such rule: tripartite information must be negative. Think of it as monogamy of correlation—if region A is highly entangled with regions B and C together, it can't be equally entangled with each separately.
For ordinary quantum systems, tripartite information can swing positive or negative. For holographic systems, negativity is mandatory.
Enter the sparse SYK model. To reduce computational cost, researchers randomly delete terms from the full Hamiltonian, retaining only a fraction p. When p drops low enough—say, keeping only O(N) terms instead of O(N⁴)—prior work suggested the gravitational dual should vanish. Chaos indicators change. Spectral statistics shift from the Wigner distribution characteristic of quantum chaos to a Poisson distribution signaling integrable, non-chaotic dynamics.
The team behind this study asked: does entanglement structure tell the same story?
Flavor Space, Not Outer Space
Standard entanglement entropy measures how much information is shared between spatial regions. But the SYK model has no space. Instead, researchers partition the system by flavor—grouping different species of fermions and asking how entangled one subset is with another.
This distinction matters. There's no known gravitational interpretation for flavor entanglement. No Ryu-Takayanagi formula mapping entropy to surface area in a bulk geometry. The team wasn't testing whether sparse SYK is holographic in the usual sense. They were checking whether it satisfies the same entropy inequalities that holographic field theories must obey.
They computed tripartite information for thermal states across varying temperatures and sparseness levels, from the full model down to configurations with just O(N) terms. They extended the analysis to four- and five-party inequalities. Every holographic entropy inequality they tested—subadditivity, monogamy, and the more exotic constraints discovered for five or more regions—held.
Always. Regardless of sparseness.
Even when other probes declared the model non-holographic.
Tripartite Information as a Weak Signal
Why does this happen? The paper offers numerical evidence, not a proof, but the pattern is clear. When researchers plot tripartite information against purity (a measure of how mixed the quantum state is), curves for different sparseness values nearly overlap. Sparseness shifts the range of accessible purity—sparser models have higher ground-state degeneracy, limiting how pure the system can become—but within that range, the entanglement structure barely changes.
The team contrasts this with two non-random vector models, simple systems where fermions interact without randomness. There, the sign of tripartite information depends sensitively on how you partition the system. Choose one grouping, and it's positive. Rearrange the boundaries, and it flips negative. The entropy inequalities shatter.
That fragility suggests negativity of tripartite information reflects something about typical entangled states in systems with many-body interactions and disorder, not necessarily holography per se. The sparse SYK ground states, even at extreme dilution, remain complicated superpositions in the partition basis. The vector model ground states are not. Typicality may be what matters.
Necessary but Not Sufficient
None of this proves sparse SYK has a gravity dual. The team is clear: satisfying holographic entropy inequalities is necessary for a field theory to admit a gravitational description. It is not sufficient.
An analogy: if you want to enter a building, you need a key. But having a key doesn't mean the building exists.
Other diagnostics—spectral form factors, Lyapunov exponents—suggest that when the number of interaction terms drops below a threshold (roughly N terms for a system of N particles), chaos dies and holography with it. The entropy inequalities, by contrast, seem blind to this transition.
Why? Perhaps because they probe global, ensemble-averaged properties rather than fine dynamical details. Or perhaps the connection between flavor entanglement and bulk geometry is more subtle than anyone has articulated.
The paper doesn't resolve this. It documents it.
What Comes Next
The findings open several paths. One: develop analytical tools to explain why multipartite entropy is so robust. Current derivations of holographic inequalities assume spatial entanglement and the Ryu-Takayanagi formula. A general proof applicable to quantum-mechanical models like SYK would clarify whether these constraints are fundamentally about holography or about something broader—typicality, perhaps, or the structure of strongly interacting fixed points.
Two: test higher-order inequalities. For six or more regions, the list of independent constraints explodes. The team checked a representative sample, but not the full set. A violation lurking in unexplored corners would sharply refine what these inequalities actually detect.
Three: extend the analysis to SYK clusters—lattices of SYK models with inter-site couplings that restore spatial structure. There, spatial and flavor entanglement coexist. Do the inequalities behave differently? Does the emergence of bulk geometry correlate with stricter entropy constraints?
And four: confront the interpretive question head-on. If a model satisfies all known holographic entropy inequalities yet fails every other holographic test, what does "holographic" mean? The term may need refinement.
For now, the result stands as a caution. Entanglement entropy is a powerful probe, but like any instrument, it has limits. In the sparse SYK model, those limits are now visible. The inequalities hold even when the gravity presumably doesn't.
That's not a failure of the mathematics. It's an invitation to understand what, exactly, the mathematics is telling us.
Credit & Disclaimer: This article is a popular science summary written to make peer-reviewed research accessible to a broad audience. All scientific facts, findings, and conclusions presented here are drawn directly and accurately from the original research paper. Readers are strongly encouraged to consult the full research article for complete data, methodologies, and scientific detail. The article can be accessed through https://doi.org/10.1007/JHEP04(2025)194






