This paper studies the complexity of the graph homomorphism problem in terms of the degeneracy of the target graph and shows that under ETH, there is no efficient algorithm for this problem.

This paper proposes two algorithms for approximating the number of four-cycles in graph edge streams, improving upon existing theoretical bounds and handling non-concentrated four-cycles in one-pass.

The paper shows that an oddomorphism between graphs does not imply graph minor containment and introduces a new structural relation called split-off minor.

This paper introduces DDC, a Dead-Direction Conditioner that keeps a deep network's optimization on the symmetry quotient by conditioning the optimizer's state in the orbit decomposition of a $G$-inva…

This paper provides the first integrated analysis of model dememorization, unifying unlearnability and unlearning methods, and offering theoretical guarantees on dememorization depth.

This paper establishes a Coding Theorem in the context of symmetry groups and develops a connection between subgroups of a group and subsets of binary strings.

This paper establishes mathematical limits of AGI safety, proving structural unverifiability as the core barrier.

This paper introduces normalized persistence, a version of persistent homology in topological data analysis, and proves its quantum hardness under the standard assumption that DQC1 is not in BPP.

This paper investigates the closeness of singularity for $n imes n$ unimodular matrices, specifically for $(2k+1)$-ary and $(k+1)$-ary cases of $4 imes 4$ nonnegative unimodular matrices.

The paper introduces Decaf, a system that uses automatic feedback and search to significantly improve the semantic correctness and accuracy of neural decompilers, boosting the decompilation rate from…

The paper introduces HOPE, a mathematical framework for network compression that shifts representation deconstruction from the discrete domain to a Hilbert space, enabling unbiased architectural decis…

The paper introduces DEBENCH, a novel framework that evaluates binary decompilers based on three orthogonal dimensions—readability, recompilability, and functionality—revealing that functional recover…

GJDNet proposes a joint disentanglement framework to enhance the robustness of Graph Neural Networks against adversarial attacks by simultaneously stabilizing node representations and decision boundar…

The paper proposes a hashing-based framework using Differential Privacy to generate and release private datastores for retrieval-augmented AI systems, achieving strong privacy with minimal accuracy lo…

This paper uses ergodic theory to study statistical properties of smooth sequences over the odd alphabet {1,3}, defining a notion of type for those sequences and proving unique ergodicity for subshift…