The authors introduce IVD-SSM, a model for narrative similarity that uses a hybrid State-Space Model and a Structurally Gated Alignment head to disentangle structural invariants from lexical variants.

This paper evaluates the causal reasoning abilities of large language models and finds that they rely heavily on lexical pattern matching rather than structural reasoning.

The paper proposes Dual-Scale Retentive Dynamics (DSRD), a unified framework that improves representation learning on dynamic graphs by jointly modeling evolving temporal and structural dependencies.

This paper introduces a new approach for scalable causal discovery in directed acyclic graphs (DAGs) with clustered data and local cluster-level effects.

This paper presents a method for identifying probabilistic structures from empirical probability tensors using algebraic statistics and Kronecker-stack class of configuration matrices.

The paper proposes a novel temporal and structural credit assignment framework to efficiently optimize multi-agent LLM systems by decomposing the error signal and using targeted, discrete gradient upd…

The paper theoretically analyzes the properties that optimal sparse autoencoder (SAE) dictionaries must satisfy, deriving constraints that explain observed SAE behaviors like hierarchical splitting an…

This paper introduces StructureClaw, an artifact-centered workbench for evaluating structural-engineering agents, and presents StructureClaw-Bench, an executable benchmark for testing these agents.

This paper introduces GraphSteal, an attack framework demonstrating that Graph RAG systems can leak substantial portions of a hidden knowledge graph by treating them as structural oracles.

This paper proposes a novel estimator for the target linear coefficient in a partial linear model with black-box nuisance estimation and establishes its unimprovable error rate.

The paper introduces the Vector Network (VN), a novel recurrent architecture that replaces fixed weight matrices with reusable weight atoms, enabling superior compositional generalization by making st…

This paper analyzes the multilinguality of LLMs by examining their structural properties, finding that low-resource languages are structurally more distinct from English than high-resource languages,…

This paper proposes Hierarchical Block-Local Learning (HBLL), a framework for training deep neural networks without full end-to-end backpropagation, achieving $\mathcal{O}(\log N)$ parallel time compl…

The paper formally models structure-informed multiple sequence alignment (MSA-S) as an NP-complete optimization problem, establishing a strong computational complexity baseline for the field.

The paper introduces PRAXIS, a novel algorithm that efficiently approximates the computation of 'Rashomon sets' for decision trees, significantly reducing memory and runtime complexity.

The paper introduces Semantic Triplet Restoration (STR), a novel protocol that converts complex table structures into atomic semantic triplets, improving table question answering by providing explicit…

This paper presents the benefits of visual pretraining for foundation model intelligence, outperforming text-only pretraining on multiple backbones and benchmarks.

The paper introduces TN-SHAP-G, a novel framework that uses graph-structured tensor networks to efficiently approximate and compute Shapley values and interaction indices for black-box models, overcom…

The paper proposes SubFit, a novel compression technique that achieves superior LLM compression by replacing non-contiguous, submodule-level components (Attention and FeedForward) with lightweight res…